| last updated:29 Dec 2003 16:12 UK time |
 |
| JOS Statistics - Recent Comments (Comments added for week ending Sun 28 Dec 2003) | View Other Weeks |
| pseudo-RAID on a single drive? | Sun 28 Dec | Alex.ro |
| I understand RAID uses two drives in parallel to double the throughput. Whats to stop a single drive being used that way? After all, any drive has more than one head, and just using the two heads on the same platter in parallel would double the throughput. Not to mention using four heads, and so on. Why has this never been done? Is this a bandwidth limitation in the hard drive controller? |
| Sun 28 Dec | Mitch & Murray (from downtown) | Wow ... |
| Sun 28 Dec | MX | This question is discussed in depth at http://www.storagereview.com/ I think it's in the FAQ. |
| Sun 28 Dec | MX | The FAQ question 'Why has no one made a drive with internal RAID 0?' is answered at: http://www.storagereview.com/php/tiki/tiki-index.php?page=InternalRAID0 |
| Sun 28 Dec | Philo | Hm. They don't mention the first reason I thought of: A single drive cannot, by definition, be a RAID - every single word in the acronym defines more than a single drive. Having said that, then the other reason for trying to 'RAIDify' a single drive would be to deliver performance, and obviously drive manufacturers think there are better ways of doing that (SATA, increased drive velocity, increased data density) Philo |
| Sun 28 Dec | Mark Smith | I know that this is totally out of left field... Back in the day, Netware beat the pants off of all other NOSes because of its drive geometry aware disk device driver. The idea was that Netware would sort sector requests in such a way as to minimize drive head movement and maximize the number of platters involved. While this might mean that the sectors asked for might be returned in a different order, the overall performance was *much* better. This sounds akin to what an internal RAID would accomplish for performance. On modern hardware, it's almost impossible to achieve these kind of performance increases because just about all drives lie about their geometry. |
| Sun 28 Dec | MX | This feature of Netware was called elevator seeking, right? |
| Sun 28 Dec | Mark Smith | Yup, I think so. |
| Mem | Sun 28 Dec | GP |
| Dont know if its the place to ask this, but... Does anybody know what this is ? http://joelonsoftware.com/mem.html |
| Sun 28 Dec | Brad Wilson | The name of the people killed in the 9/11 attack (not sure if it's just New York, or includes Washington and Pennsylvania, too). |
| Sun 28 Dec | no name | 9/11 |
| Sun 28 Dec | GP | OK, in fact I had guessed just after asking. I should sometimes think twice. |
| Sun 28 Dec | Snoop | Hey dude, you don't know, you ask then you find out fo' shizzle. Peace! |
| "IT shifting away from Microsoft"? | Sun 28 Dec | Mitch & Murray (from downtown) |
| Wonder if anyone saw this link on Slashdot and has comments on the article referenced: http://www.theinquirer.net/?article=13350 Basically it postulates that the IT industry is shifting, albeit slowly, away from Microsoft. I have no axe to grind with Microsoft nor am I a Linux cheerleader, but IMHO the article may be drawing a reasonable conclusion. Comments? |
| Sun 28 Dec | Mark Smith | I have noticed an increase in the heterogeneity of large corporate IT shops; 2 years ago you could pick companies off of the Fortune 500 that had a stated Microsoft-centric IT policy. That doesn't appear to be the case anymore. I also do think that MS has more high profile diificulties now than ever before. XBox is a money loser (and battling with Nintendo for *distant* 2nd place), .Net adoption has clearly not proceeded anywhere near the rate MS planned (espcially considering the hype), new versions of Office aren't an automatic upgrade, there's nothing exciting coming down the pike for a long time (Longhorn, Yukon, etc...) and MS blamed security issues are a monthly topic. MS isn't doing anything exciting right now while there's lots of excitement in the Linux community. I can see how the author could relate the two but I don't think that's entirely accurate or fair. For one thing, I don't think that the author entirely understands the corporate volume licensing program. Clearly, there are many corporations that are staying put with software versions and not signing new agreements (i.e. Gartner reports that commercial deployments of WinXP for organizations with 10k+ employees is 6.6%). That definently has an impact. In terms of Wall Street reaction, the fact that MS is having a flat quarter is an important thing to keep an eye on. MS has been accused in the past of shaving revenue off of fat quarters to prop up lean quarters; I believe there was some creative accounting that went on in the past that doesn't go on anymore. Right now, a flat quarter can be written off (economic slow down, IT spending, etc...). However, I believe that in the near future MS will significantly miss analysts estimates and there won't be a good excuse (espcially if firms like Intel and IBM are doing well). Since MS is the company with metrics that always go up, I could see there being something of a panic around the company's future (bad terms like 'market saturation' and 'lack of innovation' come to mind), a slew of downgrades, institutional ownership pull back, and an employee morale hit (since they're so loose with option and stock grants). That will stall MS's market momentum and that's when things could get interesting. |
| Sun 28 Dec | no name | 'MS isn't doing anything exciting right now' what about longhorn, .net, winfx, xaml... |
| Sun 28 Dec | Almost Anonymous | 'what about longhorn, .net, winfx, xaml...' *yawn* |
| Sun 28 Dec | Philo | 'i.e. Gartner reports that commercial deployments of WinXP for organizations with 10k+ employees is 6.6%' That's a silly analysis. For corporations, it's tough to argue WinXP over Win2k. A bigger indicator will be adoption of Win2k3 in the server market. As for the article, well, it *is* the Register, and let me sum up the author's agenda with this one quote: 'Somehow, people didn't buy the fact that $1,000 a head was cheaper than free' If he can't even make a comparative mention of TCO (if you want to argue Linux's TCO is less than Windows, fine, but it ain't ever free), then IMHO that calls the entire article into question. Just my $.02 Philo (obviously, this is solely my opinion and does not reflect the views of Microsoft) |
| Sun 28 Dec | Ankur | There ain't so much exciting going on in Linux right now either. |
| Sun 28 Dec | ...or use a real database... | Why sure there is! In fact, I think MySQL just got transactions a few months ago.... |
| Sun 28 Dec | Shlomi Fish | Well, the article had some flaws, but as a general rule, the trend is there and probably growing. Linux is not exactly the hard to use, guru-friendly that it was some years ago, and it's pretty usable for most clueful or semi-clueful people. (even if they don't know how to use the command line). Not all the software there is up to par or as complete as the equivalent Windows offering. Naturally, there are random problems that may be encountered if you push the system to its limits, but so there are in Windows. (my friend has Windows XP and he can't seem to run Mozilla there for some reason) Whether Microsoft will survive the hole they dug for themselves or not, is hard to tell. What I think will happen is that if it survives in one piece, it will be just one company among many who ships cross-platform UNIX and Linux friendly products. I think there's still room for innovation and software, and Microsoft are known for improving the usability and feature-sets of products to perfection. This skill is still needed. |
| Sun 28 Dec | Walter Rumsby | I couldn't be bothered reading the whole thing. I found the author's tone childish. Anytime anyone holds up Netscape as an example of a company that 'evil Microsoft killed' I roll my eyes - Netscape's incompotence and poor products killed Netscape; in fact most of the companies Microsoft 'killed' really killed themselves through stupidity, unrealistic pricing and poor product quality. I do think the environment is becoming more hetrogenous and this presents a problem for Microsoft. A lot of people seem very wary of Mono, so unless Redmond takes a different tack they're not going to be as ubiquitous as they have been. Personally I don't think that's a big problem - diversity is beneficial. Still, no one is really trying to compete on the desktop (I say this as an OS X user) so most of us will keep seeing that Windows flag every day. |
| Sun 28 Dec | fool for python | 'MS isn't doing anything exciting right now' 'what about longhorn, .net, winfx, xaml...' The technologies are fine, the problem is with the 'right now'. .net is right now. Lonhorn is *years* away. Monopolies have a habit of pre-emptive anouncement designed to freeze the mnarket while the products are developed (because they can). If the current products security were good enough, they might wait. |
| Sun 28 Dec | Scot Doyle | Philo has a good point when he mentions the traditional bias of The Register. It seems to me that the biggest problem MS has is the fact that Longhorn is so far away... not only in release, but even more in market penetration. Let's say it has a significantly better UI. That is great, but what is going to make people upgrade to it in sufficient numbers to create a market for it? The only thing of interest I see happening for MS over the next few years is the penetration of .NET into the market... and I don't see how that is going to help them in terms of revenue. |
| Sun 28 Dec | Philo | Yeah, only monopolies do that. Philo |
| Sun 28 Dec | dmann | IMHO Microsoft's strategy is to be the _only_ company with a major propriety IT-industry standard. That means maintaining the position Windows and Office have and preventing for example Playstation or Symbian from evolving into similar cash cows as the core MS products. I doubt Bill worries too much about Xbox profits as long as it keeps Sony in check. Now can MS maintain it's strategy ? my guess would be no. The reason is simple, there is a huge demand in the market for an MS tech alternative, if not for any other reason than to push prices lower and the general dissatisfaction with the Microsoft Tax. It will take a while and MS will probably keep most of the market but that is still very different from the current 'monopoly' position, and it will hit the bottom line. That said I saw one MS guy (from DirectX team) last month and he was pretty much a genius. I expect very interesting stuff from these boys with big brains and wallets. -dmann |
| Sun 28 Dec | Frederic Faure | BTW, the article is published at the Inquirer, not the Register :-) |
| Sun 28 Dec | no name | The article is a typical Linux cheer squad wish list. Big deal. As with everything Linux, it pays to look to the fundamentals. |
| Sun 28 Dec | Dennis Forbes | 'The article is a typical Linux cheer squad wish list' The article is pure FUD. Indeed, my Slashdot account (yes, I fear to say that I do have a Slashdot account...a 4 digit account no less!) currently has a score:5 post in that discussion deriding the conclusions of the ridiculous article. Having said that, for whatever reason (be it people not having a convincing reason to upgrade, market saturation, etc) Microsoft revenue has flattened and even declined -- This _is_ unprecedented for Microsoft, and in the ramp up to this Microsoft started the whole activation push to try to increase revenue by decreasing piracy (of course the reality is that a lot of the pirates just installed Linux), and by formulated some extremely hostile licensing plans and offensive stances towards business (like forcing you to buy a desktop license for every machine in your organization even if you're buying them pre-loaded from Dell). I expect much more of the same, or worse, as a bit of a desperate campaign to justify the market capitalization continues, as many of the other ventures Microsoft has entered into have been disasterous. |
| Sun 28 Dec | soothsayer | Of course most of you are coming from US, therefore żou are accustomed to US-centric views of anything, and choose to neglect the growing Linux usage in the other part of the world. Side note: in some countries the only way for MS not having to be replaced by Linux in desktop is not to enforce software piracy. |
| Sun 28 Dec | Skeptic | I agree that the article was pure FUD from a questionable source. Yes it is The Inquirer not The Register, but they were both founded by Michael Magee and are philosophically one and the same. Also, the fact that MSN did report a profit last month and a few other picayune details indicated to me that the writer didn't do his research or put any objective thought into the article. Anyway, opinions aside, just where is Microsoft really losing out to Linux? Their 2 primary sources of profit are the Office Suite and operating systems. In most home and desktop office environments I don't see them losing out to Linux on either of those two fronts. On the server side, where Linux is growing, are they losing business to Linux or is Linux cannabalizing other *nixes? |
| Sun 28 Dec | David Roper | IT shifting away from Microsoft? Absolutely, and not because of any trivial differences between Windows (what ever flavour) and any other OS, but because of global demography. I think most people who post to this forum simply do not comprehend the balance between the US and Western European (which are the key markets for ALL industrial products) and Indo-China. The latter is by far and away the greatest future market, whether for industrial or domestic products or for services. On the balance of probabilities, my estimation is that they will not choose Microsoft (nor Sun, nor Oracle, nor SAP, nor Siebel, nor...). This is not to say that Microsoft (and the others) will not remain a major play in its current (quasi-monopolised) markets, but for a variety of reasons - some economic but mostly social and even more importantly political - I think it will be very hard put to repeat its US/ European penetration on a truly global scale. |
| Sun 28 Dec | Mark Smith | It seems like Linux is eating into the low-mid commodity server share from both MS and other UNIX vendors. It's hard to get reliable stats on servers; you really need to look at work done/transactions processed. Check out the latest Netcraft web server survey: http://news.netcraft.com/archives/web_server_survey.html You can see from this that IIS peeked around the middle of 2002 and has been losing share ever since. Web serving is definently a commodity and it's clear that Apache is the big winner in that space (and Apache on Linux is far and away the post popular configuration). Now, I realize that this is an apple to oranges comparision (web server vs. all servers) and that there are problems with the way that Netcraft does the survey (web parking skews the results). Nonetheless, there's something to take away from this. Mid size business espcially (200-1200 employees) seem to be flocking to Linux for file serving, network services (DHCP, DNS), web serving, and e-mail. Basically, anywhere that isn't overly driven by 'strategic thinking' seems to be deploying commodity services on Linux instead of the MS Small Business Server type packages. |
| Sun 28 Dec | Philo | Given two installations: 1) MSN, with a few dozen web servers in a farm 2) An ISP with one 4-way box running Apache serving a few dozen picayune websites How does Netcraft count those? Philo |
| Sun 28 Dec | max | Basic 'web serving' has became a commodity. So, Microsoft should offer IIS for free, with every version of Windows they ship. They already do that. They need to work on making IIS a lot easier to configure, and also more secure. And they also need to offer free Windows to educational institutions, and maybe a Windows Free Edition, which will only act as a limited web server. |
| Sun 28 Dec | Walter Rumsby | They need to let me use my X-Box as a CHEAP (Java capable :)) web server. Please. |
| Sun 28 Dec | Michael Koziarski | Philo, The netcraft survey will affect MSFTs share in the same way. I remember reading a while back that in an effort to work around this, some other outfit surveyed just fortune 1000 companies and the percentages were almost identical. I'll dig around for the link. Cheers, Koz |
| Sun 28 Dec | Mark Smith | Philo: Netcraft is looking at domains. So, 1 server that hosts 100 domains would count 100 times. 25 servers in a round robin configuration supporting 1 domain are counted once. Netcraft also doesn't take into account that http://www.ebay.com is more significant in terms of workload than http://www.aiworks.com. |
| Sun 28 Dec | JWA | I have a few comments/questions: I totally didn't follow this comment: 'A lot of people seem very wary of Mono, so unless Redmond takes a different tack they're not going to be as ubiquitous as they have been.' What does that mean? My first blush was that you were referring to the Mono Project, the .NET open source implementation. But that doesn't make sense in the context of your sentence, so maybe you meant it as a shorted Monopoly? What is this FUD I suddenly see everywhere. I've been able to figure out every other weird net-speak acronym, but not this one. Philo getting a job at MS has bummed me out. I've always enjoyed and respected his (your) comments, and I was happy to hear about your cool job. Now, though, it seems as though every one of your posts are defending MS or cheerleading for them and their products. This is totally understandable as this is now your family, and it's true that I agree with everything you've said both technically and in personal opinion. However I just don't read the comments as being completely objective any more. I've never really noticed this phenomenon before and it's interesting to me. --Josh |
| Sun 28 Dec | JWA | Ok, Googled like I should have and result #1 is: http://www.geocities.com/SiliconValley/Hills/9267/fuddef.html First line on the page says: 'FUD stands for Fear, Uncertainty, Doubt'. Got it. --Josh |
| Sun 28 Dec | Philo | Josh - for what it's worth, it's bummed me out as well. Working as a salesbeing for MS is interesting, because I truly believe in their products. If you'll look over my posts since I've been posting to JoS, in general I've been a staunch supporter of SQL vs. Oracle or MySQL, .Net, Windows vs. Linux, etc. I personally believe that in each case the MS platform has been made easy enough to maintain that it allowed me as a developer to concentrate on the work I had to do instead of being an admin. The joy of the MS job for me is doing something I truly believe in. Unfortunately, I'm sure my employment will have that tainting effect on what I write now - people will think 'oh it's the Microsoft shill again' My writing's been off-center lately as I've tried to come to terms with my new position, what company policies are about posting, and how to bring it up on here. Obviously pro-MS stuff (esp. in the normal course of discussion, as opposed to in a 'which is better' bunfight) was easier to start on. (An interesting side effect of the job is that some products are even more interesting to me now as I get exposed to a lot of their capabilities I wouldn't have paid attention to before. Thus the new love of Office 2003...) ...and I won't be posting MS vents as much, as I'm more likely to go hunt down the person responsible, now that I can. [grin] In summation - I haven't sold my soul, I'm still trying to speak candidly, and I still, sorry to say, happen to like all of Microsoft's stuff. Philo |
| Sun 28 Dec | JWA | Philo, I understand completely, and I agree that you haven't changed what you're saying (though I've noticed the deeper exposure to other apps). It's just a weird phenomenon that I noticed and for some reason mentioned. By the way, your statement that 'it allowed me as a developer to concentrate on the work I had to do instead of being an admin' is spot-on. Have a good evening, --Josh |
| Sun 28 Dec | Brazilian Dude | LOL, loved the term "Linux cheerleader". |
| Sun 28 Dec | Walter Rumsby | Sentence 1: 'I do think the environment is becoming more hetrogenous and this presents a problem for Microsoft.' The 'new' computing environment is not just Microsoft. I guess this doesn't require further analysis. Sentence 2: 'A lot of people seem very wary of Mono, so unless Redmond takes a different tack they're not going to be as ubiquitous as they have been.' Mono would theoretically allow .NET to run on a Linux (or OS X, or PS2, or ...) box, so while MS might lose licensing $ from people not buying Windows, those people could still use Microsoft applications and frameworks written for .NET running under Mono (and theoretically MS could potentially implement everything so that it runs in the CLR - e.g. SQL Server [or at least something like JET], Visual Studio, ASP.NET, etc). This still earns Microsoft dollars, and doesn't lock them out of consideration for *nix machines (right now you're not going to be deploying SQL Server on your nice new Linux server). Most importantly it keeps users doing things 'the Microsoft way' - users will still look to Microsoft for direction. What I was thinking when I mentioned Mono was that Mono seems to offer a way of embracing the 'new' hetregenity of the corporate computing environment whilst still doing things the MS way (as outlined above). However (and it is probably just FUD) the majority of discussion of Mono I see seems to say something along the lines of 'if Mono ever takes off MS will kill it'. (Somebody, I think it was Dennis Forbes... or perhaps Atkins..., posted something positive here about their view of Mono and that really stands out for me because his perspective was different to almost everything else I've heard on Mono). The general attitude people have towards Mono is important because while people think Mono has no future and if they are planning to use Linux in their organisation, .NET-based development becomes a lot less attractive (sure they could use web services, but once you make that decision you're putting on the ugly suit right there). Hope that clears things up. |
| Simple ASP Question | Sun 28 Dec | Chi Lambda |
| Im pulling PRODUCT_SPECS from my database into a large Recordset and displaying them on my site. Within the specs information sometimes there are titles of products. Desire: Id like to automatically convert each product title found in the specs to be a link to another page. I figure to open another Recordset from the PRODUCT_TITLES table and compare that listing to PRODUCT_SPECS and where a title exists make it into a link. Problem: Some product titles actually contain other titles. Example: - TurboCAM Milling (software) - CAM Milling (machine) Question: Whats the best and simplest way to compare the information and create an automatic link? And how do I make the code differentiate between text found within other text and that which appears alone? Thanks for the help. |
| Sun 28 Dec | Matthew Lock | So we can help post the fields of the product_title and product_specs tables. |
| Sun 28 Dec | dmooney | I'm not sure why you wouldn't have a numeric primary key that you could compare with. I'm also not sure why you wouldn't use join the tables using SQL, even if you did have text keys. |
| Sun 28 Dec | Chi Lambda | Note: the product titles found in the specs do not come from the PRODUCT_TITLES table. |
| Sun 28 Dec | the artist formerly known as prince | hmmm sunday + simple question = Homework? |
| Serving ads in software | Sun 28 Dec | Alex |
| I am developing a piece of shareware software. Can anyone please advise how i may include ads in it? Most of the sites i have been to only server banner ads. How does flashget or kazaa do it? thanks alex |
| Sun 28 Dec | Matthew Lock | Does ad supported shareware actually make any money? I know that a lot of shareware have removed their ads. (Read this http://www.getright.com/statement.html) It's also the quickest way to get your shareware labeled spyware. |
| Sun 28 Dec | Elan Dekel | You might want to check out: http://www.cydoor.com/ Not that I recommend it or anything, I just know that they seem to be making an awful lot of money doing this stuff... |
| Sun 28 Dec | Matthew Lock | Cydoor is exactly what I meant about getting your software labeled as spyware. Check out the pages returned when you search on 'cydoor' http://www.google.com.au/search?q=cydoor |
| Sun 28 Dec | www.marktaw.com | Opera does this and their documentation is pretty good. Banner Advertising Implementation http://www.opera.com/docs/ads/index.dml Advertise In Opera http://www.opera.com/advertise/ You can probably learn a lot from these two links. |
| Sun 28 Dec | Frederic Faure | Ads are anoying. Maybe you'll have more luck making the basic version free, but providing an enhanced but commercial version. |
| Sun 28 Dec | Dennis Forbes | Speaking of ads, I have been quite surprized that there aren't "sponsored" applications -- not banner ads rotating downloaded ads or anything like that, but rather just a particular corporation branding -- i.e. you load up SmartFTP and the splash screen proclaims "Brought to you by coke", and maybe the color theme is red and white with minimal coke branding. Given the proliferation of branding in the rest of the world, the fact that many very popular apps aren't branded seems bizarre to me. I'm, of course, talking about free versions of software -- this does not carry over to pay products. Speaking of that, the wife bought the "My Big Fat Greek Wedding" DVD - Unbelievably there are literally 15 ads before the movie starts. They are all ads by the same company promoting their other garbage movies, but this abuse of customers just blew me away. |
| Sun 28 Dec | Jason | >> I have been quite surprized that there >>aren't 'sponsored' applications -- not banner ads >>rotating downloaded ads or anything like that, but >>rather just a particular corporation branding The Weatherbug http://www.weatherbug.com/aws/index.asp does this. If you use their free version, every so often you pick a sponsor from a list of 20-30 companies and the background of the application becomes a billboard for that company. Jason |
| Sun 28 Dec | Alex.ro | >> Ads are anoying. Maybe you'll have more luck making the >> basic version free I agree. That's why I quit using Opera. Winamp is free as in beer, but they have a paid version with some extra functionality, so they get both ample installed base, and revenue. If you go for the ads, at the very *least*, make sure your ads aren't *animated*. |
| Sun 28 Dec | max | This is BS, in my opinion. The makers of WinAmp are called Nullsoft. Nullsoft was purchased by AOL. So - AOL finances the whole thing - you know how big US companies are - they can afford to drop some money on such a thing. |
| Open source complexity issue | Sun 28 Dec | Mike |
| Ive snipped this from another post on a different topic here. It stood out like a sore thumb. Since this is turning into a discussion of how open source will affect us, let me recommend that everyone read a great article by Marc Fleury. Basically he explains why he CAN and IS making money by providing an open source project such as jboss. No small time ISV is going to be able to come up with a complex Application server (or database program, or operating system, or ...) and make money doing it. BUT, many ISVs can and will make money writing custom software FOR those big, complex applications. Hence the reason OPEN SOURCE WILL MAKE US ALL MONEY.http://www.jboss.org/modules/html/white3.pdf So are we to believe that this is an incentive for OSS to write complex code so they can have hope one in hell of selling documentation and services and so ISVs can help you wade through the muck. If OSS goes this route to make money, that is a bad thing. Although thinking about it, IBM global services springs to mind. I wonder why? |
| Sun 28 Dec | John Rose | I don't see how that's a lot different than an ISV making money off of solutions built on closed-source commercial software like IIS or Oracle. Just because a company can plunk down $___ for Oracle doesn't mean they understand it, just because it comes with manuals and support. They still need to pay skilled people (whether employees/contractors/ISV's/etc) to make sense of it and built on top of it. In terms of 'how can an ISV make money?' I don't think the closed/open source debate is really applicable, except for the costs of the software itself, which is usually a relatively small cost of the overall development effort anyway. |
| Sun 28 Dec | Matthew Lock | The fact that Tim Berners-Lee made the world wide web an open standard certainly made us developers a lot of money. |
| Sun 28 Dec | John | Please look at income and profits of software companies based on commercial software, and at income and profits of software companies based on open source. This information is easily accessible for public companies - you can find it using a search engine. You will see that: - lots of commercial software companies are profitable (even if we take giants like MS and Oracle out of the equation) - very, very few OSS companies are profitable - the OSS companies that are profitable have very low profit and very low income compared to commercial software companies, and thus, can hire fewer programmers For example, fans cheered when Red Hat became profitable in the last quarter or so. But the Red Hat profit was very low, and it's doing much worse than commercial software companies of comparable size. Very few companies can succeed if they sell open source software. I belive that in the long term, open source creates a 'rock star-like' business environment. In the music world, a handful of people are rock stars, and make decent money, while the rest (that is, 98%) barely make a living and have to struggle very hard. In the brave new OSS world, a handful of programmers and 2-3 companies will be rich, while the rest of us will barely scrape by because every piece of software will be free as in beer and free as in freedom. You many not know this, but some open source software companies have simply resorted to begging in order not to tank. Mandrake is a recent example - they cried for help, and asked people to donate. Excuse me, but is this the future we want? Do you want to work for a software company that is in constant danger of tanking, and has to beg people to donate, or do you want to work for a software company which sells it's products, and has a healthy profit? Where do you think you would be happier? Which company do you think will hire more programmers? |
| Sun 28 Dec | T. Norman | There is a big difference between a company whose core products are open source, and companies that use open source to build commercial products or custom solutions or enhance their other products (like IBM selling mainframes that run Linux, or Joel using PHP for FogBugz on *nix). I wouldn't want to work for the first type, but there is plenty of money being made in the other type. |
| Sun 28 Dec | >> TD << | I agree with T Norman. If you want to be profitable you've got to add value to OSS products |
| Sun 28 Dec | no name | The part that sticks out like a sore thumb to me is that to make money off of open source you need to be a service provider. i.e. you need to provide computer consulting or contract programming. I'd much rather be selling a product than providing consulting. Good luck trying to compete with JBoss though. |
| Sun 28 Dec | Frederic Faure | Although some people use OSS tools simply because they're as in beer, others (or the same) use them because they're free as in speech, ie. as much as possible, they'd rather use tools that do not lock them into a proprietary solution, raising the cost of switching to another solution in case the company goes under or raises, lowers the quality of its products or service, or raises its price to an unreasonable level. Business 101: Always have a plan B :-) We are then, however, faced with the issue of how to allow developers to make money off a tool that is freely available. As explained in this thread, OSS tools that require full-time employees for developing it (salaries, rent, etc.) and/or do not require consulting to make it work because it's so basic... are not company-friendly. It's not an easy problem balancing the need for individuals to make $, and those same individuals to share knowledge to improve society (where would scientific knowledge be today if each scientist kept his discoveries to himself, or patented them and forced others to shell out $ to use them?). |
| Sun 28 Dec | Kerry Heffermen | 'where would scientific knowledge be today if each scientist kept his discoveries to himself, or patented them and forced others to shell out $ to use them' Isn't that how it works right now? |
| Sun 28 Dec | Shlomi Fish | Well, I personally always felt that ASP (=Active Server Pages) was a scheme by Microsoft to provide a living for a great deal of people. I mean, I once had to write a simple form handler for IIS, and I scanned the ASP documentation for something that will enable me to send E-mail, and found absolutely nothing. I ended up writing it in Perl, where it worked perfectly. It turns out you need to buy a specialized ActiveX object that can send E-mail. Ditto for using regular expressions, doing encryption, or for opening files (!). And Microsoft does not provide all these things - other vendors do. In Perl you can find almost anything on CPAN, and the core language itself has any system mechanism that someone bothered to write a binding for (files, sockets, users, processes, the works). Ditto for PHP (albeit getting an extension is less straightforward than with CPAN) and I think for Java as well. So they gave people a braindead solution, so there will be an entire culture of vendors springing to complement it. I'm not saying this phenomenon is limited to what Microsoft does. It exists anywhere, even in open source places. However, no software can ever be fully complete to satisfy all needs, or so it can be used out of the box. You just have to make sure it is extensible solution, so other people can build upon your efforts. This is the case for JBoss. From what I know of it JBoss is far from being braindead or incomplete. But many firms who would wish to deploy it (or Perl for that matter) would need to customize it or otherwise implement things above it that don't come by default. It makes no difference whether it is open source or not - that will always be the case. |
| Sun 28 Dec | Shlomi Fish | John: for your information, most of the commercial software startups fail miserably as well. As for the profitability margins: you may be correct. If you hit the jackpot providing a (usually very costy) commercial solution, you can become very profitable, at least until an equivalent open source solution or a better commercial ones comes along and eat your market. BTW, profits do not finance programmers, they just finance new programmers and expansion. In fact, the salaries and other costs involved in hiring programmers is deducted from the total earnings, and after more deductions eventually yield the cost. Of course, in my opinion, at least, open source is more fun and more rewarding. A Freecell solving program I developed would not have had half the features it has now, had I did not make it Public Domain from the beginning. In a way I felt that I did not cheat the community by hiding the source from them or demanding money. Note that I would not mind getting any job, whether in commercial, open source or simply in-house, embedded or whatever software. I just love to program. |
| Sun 28 Dec | no name | >> 'Well, I personally always felt that ASP (=Active Server Pages) was a scheme by Microsoft to provide a living for a great deal of people. I mean, I once had to write a simple form handler for IIS, and I scanned the ASP documentation for something that will enable me to send E-mail, and found absolutely nothing. I ended up writing it in Perl, where it worked perfectly.' You're pretty much a god damn mother fucking stupid piece of shit bastard. Ever hear of CDO. I think you need to shut the fuck up or get the fuck out of the business. Oh yea, I almost forgot mother fucker, learn how to fucking read. You fucking nit wit, you just wanted to find some way to bash Microsoft. You're a complete mother fucking moron. Hence you're retarded name. From now on I'll call you Stupid Fuck. |
| Sun 28 Dec | Dennis Forbes | Good day Shlomi 'It turns out you need to buy a specialized ActiveX object that can send E-mail. Ditto for using regular expressions, doing encryption, or for opening files (!). And Microsoft does not provide all these things - other vendors do.' Regular expressions have been a part of the script interpreter engine for many years now. Opening files (fso) and sending email (CDO, MAPI) have been available pretty much from day one. Encryption is functionality exposed through the CryptoAPI available from 'Win32', but as ASP can't call Win32 functions you do need to fire up Visual Studio, create a basic COM wrapper, and there you go--encryption from ASP. Indeed, COM (via dispatch interfaces) was the ultimate interoperation mechanism in ASP and if you could expose some functionality to COM, you could use it from ASP. So in other words I entirely disagree with you -- Microsoft stated 'here's ASP, and it can talk to anything that knows how to talk via a dispatch interface', and anyone, either closed or open source, could freely release functional 'plug-ins' to be used from your code (and there are tens of thousand of free and commercial products), or alternately you could easily expose your own functionality. Your misunderstanding is quite common, however, and it derives from people looking in the VBScript documentation and looking for the library calls, when instead it isn't a reinvention of the wheel in the VBScript world, but rather is the entire universe of COM components. |
| Sun 28 Dec | Dennis Forbes | As a sidenote, and to demonstrate this point, a friend, whose site is based in ASP (they haven't made the .NET migration), needed a component to pull a thumbnail out of an uploaded video file. Knowing that the .NET Framework has fantastic graphical capabilities, and knowing some DirectX library calls to open a video file, I quickly threw together a .NET component that would pull the image file, resize it to the desired size, and save it in the desired format. Sticking a Com Callable Wrapper on the component and voila, from ASP some pretty advanced image capabilities (I even included multiple layers with various transparency, titling, etc), available via the magic of COM. |
| Sun 28 Dec | Shlomi Fish | Anonymous: what is the meaning of all this swearing? We are civilized people here and I expressed an opinion. Perhaps such that is based on dis-information I mistanekly had but nonetheless it was perfectly innocent me. I wouldn't blame Joel or any other moderator of the forum if he reomved it. |
| Sun 28 Dec | Shlomi Fish | Dennis: I stand corrected. I suppose the cost of having Visual Studio is negligible compared to the entire cost of having IIS+ASP etc., and anyhow a Windows webmaster is not as spoiled as his Linux peer and expects to pay for software. In any case, from what I heard the case was that you indeed needed such stuff to effectively worked with ASP, or at least people used them. Maybe those 'out-of-the-course' ASP winnies, cannot be expected to write a COM object by a long shot, and so would rather convince their deployer to buy something commercial. That may be a case for a great deal of PHP programmers out there as well, only with PHP you can usually get the components for free. If the documentation of ASP did not make it clear how to write a form handler that sends an E-mail message, then it's a bug in the MSDN documentation. After I converted to using Perl, I was able to find several good ways to send an E-mail, and eventually even implemented my own ad-hoc SMTP client stack there, using sockets. (that's because none of them worked on my Apache for Windows Alpha version which I installed there, due to a bug that I eventually reported and was eventually corrected - ;-)) That put aside, I'm not sure I can ever adapt to the philosophy of ASP or JSP. Their assumption is that the language is crippled, and you need to use it as a number of callbacks to either OLE objects or Java beans. In Perl/Python/PHP/etc. you don't need all that, because the core language is full-featured, and you can implement as complex logic as you want inside the script, or supporting modules. What does frustrate me is that high schools in Israel teach ASP, which requires Windows to run on. They should instead teach PHP, Perl, etc. which are cross-platform. Once I met two guys in the 'Go! Linux' Israeli conference (the first large conference of this kind), who saw that I knew my way around computers, and then asked for help with their ASP homework. I told them I only had a Win98 at home, and could not install IIS and all of the required components very efficiently. If they had a problem with PHP or whatever I would have gladly tried to help. They also were representative of one of the companies there (which sells some Linux software), and said on Linux that it was used by many computer introduers. (which is true, but that's not its primary use and value to society). And it was a Linux conference! |
| Sun 28 Dec | So give me your keys or I will say you are selfish and unreasonable | 'I felt that I did not cheat the community by hiding the source from them or demanding money' And I feel you are cheating me my not giving me your car. |
| Sun 28 Dec | Frederic Faure | Kerry Heffermen >> Isn't that how it works right now? In the private sector, yes. Not sure we're better off, though. BTW, maybe Joel should make people pay to access this knowledge they're currently getting for free on his site ;-) |
| Sun 28 Dec | Mike | 'You're pretty much a god damn mother fucking stupid piece of shit bastard. Ever hear of CDO. ' Hey Einstein, please tell us when CDO debutted. |
| Sun 28 Dec | Me | > BTW, maybe Joel should make people pay to access this knowledge they're currently getting for free on his site It takes 60 seconds to read a thread and compose a reply. That's quite different from the effort and time to develop useful software. |
| HELP! data recovery advice | Sun 28 Dec | Roose |
| I just overwrote an important file, and Im not sure if I can recover it. I was using CoolEdit to edit some important music file, and I accidentally made a bad change, so I selected REVERT TO SAVED. So you would think if I hit YES on the next dialog box, it would LOSE the changes and actually revert to saved. But NO, the way its implemented, is that it closes the file and reopens it -- and closing asks: do you want to save your changes? with YES as the default. Doh!!! What retarded design. And you cant undo after saving either, I guess. So I just did this, and didnt touch the computer, so hopefully it didnt actually overwrite the file on disk? Can anyone suggest some sort of software that will recover this? But if I install the software, theres a chance that it ITSELF will overwrite my old file, no? And hopefully leaving Windows on wont overwrite it with some background processes... and I think the swap file is statically allocated as well. Please help Im desperate, this was an important music file! |
| Sun 28 Dec | Roose | By the way, this on a Windows 2000 system with FAT32, not NTFS. thanks! |
| Sun 28 Dec | Matthew Lock | Check whatever temp directory cool edit uses. My wave editing program used to routinely crash but leave the wave in the temp directory where I could recover it. I guess I don't need to tell you to back up important work before you start editing it. ;) |
| Sun 28 Dec | Spider | Your might try the popular undelete program found at http://hccweb1.bai.ne.jp/~hcj58401/ Preferably using another computer (not the one containing the deleted file), download it and unzip it. The result is a single executable which can be run from a floppy, no install needed. This was designed to recover deleted files; depending on exactly what happened in the scenario you describe, it might help. |
| Sun 28 Dec | www.marktaw.com | Cool Edit uses lots and lots of temp files so that almost no changes are permenant. Though once you've saved the file and closed the program, I don't think those temp files will still exist. You should probably ask in an Audio forum. |
| Sun 28 Dec | anon | I can recommend OnTrack EasyRecovery for this job, it's saved me a number of times in the past. I have never used it with CoolEdit, though. http://www.ontrack.co.uk |
| Type of parser used for ASP->PHP compiler | Sat 27 Dec | Matthew Lock |
| Im wondering what kind of parsing technique was used for the Fogcreek ASP->PHP compiler. Was it a recursive decent parser or a simple code filter? |
| Sat 27 Dec | Joel Spolsky | One of these days I'll write about it. It's a real lexer, parser, ast, code generator, which is why I keep calling it a compiler. It is NOT just a simple filter (what someone called "put $ in front of every variable name and cross your fingers). |
| Sun 28 Dec | Matthew Lock | Excellent. Many so called compilers and code converters around are really just a fancy search and replace programs. |
| Sun 28 Dec | Eli Bendersky | Ahh... parsing :-) Lex (Flex) and Yacc (Bison) rule. Creating ASTs from them for well-behaving languages is fun and simple. Dumping code from ASTs when the target language needn't be optimized too much is also fun and simple. Joel, do write about it. It would definitely be a most interesting read. |
| Sun 28 Dec | Shlomi Fish | Yes, I think this article will be very welcome here. The Joel on Software web-log seems to touch a lot on software management issues (which is good), so it could use some more Computer Science-oriented articles. (like Back to Basics, for example) In any case, one should be aware that Lex and Yacc, while commonly used, are not the cutting edge of compiler design. Lex is simply a tokenizer that extract tokens based on regular expressions _without any state_. A simple Perl script can do better than that. As for Yacc - it is nice and all, and very useful. However, it too can be improved upon. One such implementation is the Scannerless Generalized LR parser: http://www.cwi.nl/htbin/sen1/twiki/bin/view/SEN1/SGLR Another is Damian Conway's Parse::RecDescent module: http://search.cpan.org/~dconway/Parse-RecDescent-1.94/ which is very powerful and capable, but kind of slow. One thing that should be remembered is that formal languages that are commonly used for parsing languages are not Turing complete. So, a custom parser is needed for some patterns. |
| phone interviews | Sat 27 Dec | anon |
| Going into my 6th interview in 3 months,I was given 3 phone interviews with 3 developers over which I was asked 2 design questions, data structure theory questions, 3 coding questions (coding line by line over the phone), over a period of 3 hours. I guess they wanted to find out if the person can code or not. How are other peoples experiences with phone interviews? Can you really judge tech skills over the phone, at least initially? |
| Sat 27 Dec | Brad Wilson | Most phone screens I've been part of used rather simple technical questions to test some aspect of the person's resume. This wasn't so much to see if they were the right fit, as to filter out a bad fit (usually from over-embelishment on their resume). What it sounds like you were put through was more like a full interview, except on the phone. Is it expensive to get you to their offices, like it's in another country? It seems pretty surprising that anybody would do an interview of that depth over the phone. But maybe with the job market the way it is today... |
| Sat 27 Dec | anon | No it's only the first stage, there's another part face to face if they find me suitable. And yes they are around a 3 hour plane ride away |
| Sat 27 Dec | Philo | By the way, for any interviewers out there, a three hour phone interview (all at once) can be tough on people. Cell/cordless batteries aside, if they don't have a speakerphone (and it's unreasonable to expect someone to have a speakerphone at home) then you're talking about some serious muscle fatigue. I'd strongly recommend breaking it into 3 1-hour interviews. You should also let the candidate know up front that's what's coming: 'you'll have two or three one-hour phone interviews, then if we like you we'll fly you in...' Just my $.02. Philo |
| Sat 27 Dec | George | I live in Bucharest, Romania. Microsoft interviewed me over the phone for about 1 hour. It was a scheduled interview - they first called and asked when it would be a good time to talk for an extended period. They asked a lot of questions - mostly design questions, and some programming questions. Then they asked me to come to the live interview, also held in Bucharest. While I wasn't familiar with their methods, I liked them a lot - they seemed to have a very 'get the job done, no useless stuff' attitude. |
| Sun 28 Dec | Evgeny Gesin /Javadesk.com/ | I inteviewed a few candidates by phone. Usually it takes up to 10 minutes, I ask 3-4 simple technical question, and, if answered, then candidates are invited to office for a more detailed interview. My 0.002$ |
| Sun 28 Dec | son of parnas | We use phone interviews as a simple technical filter. I suck at phone interviews so i don't care for them very much. |
| Sun 28 Dec | Philo | I like phone interviews when they are conversational. I *hate* the 'What is the maximum airspeed of a fully-laden swallow' bullet question type interviews. But then you all know that about me. :) Philo |
| Sun 28 Dec | Mark Pearce | Philo, Not sure about a fully-laden swallow, but this is the data for an unladen swallow: http://www.style.org/unladenswallow True for any given value of 'true' :) Regards, Mark 'Comprehensive VB .NET Debugging' http://www.apress.com/book/bookDisplay.html?bID=128 |
| Increaseing Speed of Computer - optimizing process | Sat 27 Dec | Techno-phobe |
| Hi: I have an IBM Notebook - lees than two months ole - 256 mb RAM, 1.5G Centrino. When I first got this, it ran like a charm and would boot up in a flash. Now it takes close to 8 minutes before the boot process is complte? How can I best optimize the speed? I have downloaded some software like Oracle 8i, Apache, MS Office, Zonealarm, VMware, Eclipse etc. I refraq frequently. Is there a way I can go about to speed things - I often kill the Oracle process on start up to hasten the process. I have checked the processes listed on Task manager on Google - there does not seem to be any thing nasty running. I was thinking of using Adaware or CWShredder: http://www.spywareinfo.com/~merijn/files/cwshredder.zip But was not sure if the above was any good or if they contained any spyware. |
| Sat 27 Dec | Jim | 1. Run a free online virus scan: http://housecall.trendmicro.com/housecall/start_corp.asp 2. Download and run Ad-aware 6.0 to remove spyware/adware. 3. Uninstall any un-needed apps. 4. Defragment your hard drive. 5. Reboot This should get you back to with 98% of your original computer speed at a cost of $0. |
| Sat 27 Dec | robert | Defragment your hard-disk, using either MS Defrag, or DiskKeeper. DiskKeeper can be found at: http://download.com.com/3000-2094-10224681.html Then, defrag your registry using Innovative System Optimizer. I found that this can make a lot of difference. Innovative System Optimizer can be found at: http://download.com.com/3000-2086-10252167.html?tag=lst-0-1 Finally, run Ad-Aware to remove spyware programs. Ad-Aware can be found at: http://download.com.com/3000-2144-10214379.html?tag=lst-0-1 |
| Sat 27 Dec | anon | You might also want to switch the startup attribute of you some of your services from Automatic to Manual (especially Oracle). |
| Sat 27 Dec | Rhinestone Cowboy | How can one switch processes to start manually from automatic? Thanks. |
| Sat 27 Dec | robert | Start Menu - Run type 'services.msc' then press ENTER |
| Sat 27 Dec | Brad Wilson | You may also wish to consider using Virtual PC for your development environment, and leaving your laptop more or less pristine. Then you boot into the laptop very quickly. The Virtual PCs can be kept in a suspended state. In my experience, it takes about 10 seconds to stop or start a virtual PC w/ 512MB of RAM. One downside is that your laptop would need more RAM. 512MB would probably not be enough, since the host OS will likely be taking at least 50MB for itself. |
| Sat 27 Dec | Brad Wilson | Oh, and for some reason, I thought you had 512MB. Definitely, with the stuff you have installed, you should have more than 256MB, even if you didn't run Virtual PC. I'd recommend seeing how much it would cost to bump the PC up to 1GB. |
| Sat 27 Dec | Brad Wilson | Oh, and as a final alternative, don't shut the machine down. Use Hibernation to speed the shutdown and reboot time of the box. |
| Sat 27 Dec | Jim Battin | I'd strongly recommend upgrading to 1GB of RAM. |
| Sat 27 Dec | Alex | Indeed, why not use 'sleep' instead of shutdown. It's very convenient when you need the computer for only a minute, e.g., to check the mail. |
| Sat 27 Dec | John Rose | Good suggestions so far. Start-->Run-->'msconfig' This will give you a list of what's being started up along with Windows in the Run key of the registry... you can disable each item. Upgrading to 1GB of RAM would be a nice step as well... but even without the ram upgrade, something is definitely funky with your setup... 8 minutes to boot is nooottt normal (as you've deduced). :D |
| Sat 27 Dec | MX | I have got myself a Siemens M55 mobile phone and can now check my e-mail very quickly, even on the street: I make it download the e-mail, and then browse the e-mail offline. I can do this even when driving (when I stop at a red light) or when walking on the street! This is very practical. |
| Sat 27 Dec | Alex | Hey MX, How in the world did you configure your M55? I just got one and the company staff can't configure the GRPS :( |
| Sun 28 Dec | ICBW | For a comprehensive list of all processes that are being run when you boot up, get autoruns from www.sysinternals.com. At last count there were 23 places where startup programs can be specified (not including services). You can also pick up a copy of PageDefrag there, which defrags the registry, page file and event logs. Both programs are free. |
| Sun 28 Dec | MX | Alex, I don't use GPRS. I use my phone just like dial-up - connect, do your stuff, disconnect. I'm not sure if this is called CSD, or WAP, or whatever, but the WAP browser and the e-mail client work! The reason why I don't use GPRS is that in my country, it's very expensive compared to CSD. If you want to discuss about the Siemens M55 mobile phone configuration, please use the link at the bottom of this message - no point of writing this in the forum. |
| Advice for S/W Entrepreneurs: Make More Mistakes | Sat 27 Dec | Jim |
| http://msdn.microsoft.com/library/en-us/dnsoftware/html/software12292003.asp?frame=true *** VERY INTERESTING ARTICLE *** |
| Sat 27 Dec | no name | What do you think is so interesting about it? |
| Sat 27 Dec | Jim | > What do you think is so interesting about it? Its interesting because it's true. It seems to be a different take on the ISV business. Certainly different than what JoS preaches (perfectionism, craftsmanship, mood altering office space, etc...). I mean, nothing against Joel, but who cares if the window next to your cube is big enough and has just the right view? Same goes for the color scheme of your walls and furniture. It seem's that Eric is more pragmatic and is willing to spend money where it counts, on breaking lots and lots of eggs. Learn by doing, etc.. Instead of trying to get it right the first time. Check out the guy's blog, he seems to be the anti-Joel. Similar, but different. Pretty wierd. http://software.ericsink.com/ |
| Sat 27 Dec | Alex | That's why I think the best business reading is the 'mistakes made' kind. Going over others' mistakes instead of making your own, it's priceless. |
| Sat 27 Dec | Philo | I'm not sure how you get that Eric and Joel have different approaches. In case you hadn't noticed, Joel's bionic office was the *second* office for FogCreek, and he just spent a lot of money porting one of his flagship applications to Unix. If he was the indecisive perfectionist you imply, FogCreek never would've been founded, since Joel would've refused to incorporate until he could 'do the office right,' and he would still be online trying to decide what build of Linux he should try installing first. FWIW, I dislike Eric's approach to the article, embodied in the title - the same philosophy was quoted by someone as Microsoft's philosophy of 'you're expected to fail.' What they're both doing is encouraging those who are risk-averse to take chances, but it's too negative for my taste; it's on the wrong side of the fine edge between daring and reckless abandon. But that's just me. :) Philo |
| Sat 27 Dec | no name | Some of those mistakes really are embarassing. So embarrassing it's probably a mistake to disclose them publicly. The "Investing outside our field" story is a classic sucker deal that says "I shouldn't really be in business". |
| Sat 27 Dec | I'm a fan | That is one of the most useful business articles of 2003. Well done, Eric Sink (who occassionally drops by at JOS.) Apart from his business nouse, anyone who elects himself to the hall of software Non Legends with a photo in a corn field marks himself as a special type of person. |
| Sat 27 Dec | no name | I would really like to know where he got all that money to make all those mistakes. Did it all come from sourceoffsite? Any one of those mistakes would've sunk me. |
| Sat 27 Dec | Mitch & Murray (from downtown) | I think he is taking a very well deserved poke at Joel and others like Joel whose primary marketing mechanism for their products is being an industry pundit, telling us what we are doing wrong, and 'oh, by the way, I have something you can buy from me that will make you do things better'. This doesn't make Joel a bad guy, btw. It is just his schtick. |
| Sat 27 Dec | Gwyn | What interests me about this is that the entrepreneurial people (and I've seen this before) do not succeed because they are better at things than other people; it is because they have the entrepreneurial mindset and they keep on trying, banging their heads against walls where most risk-averse people would start thinking 'whoa!' and give up. They're natural risk-takers. And because they are risk-takers that is enough to get them through to some degree of success (eventually). I'm not sure anyone who does not already have this mindset can be an entrepreneur; it's the way people are made. Reading this article was interesting, the guy's made some dreadful decisions that others (including me) would never have made and still (eventually) he's achieved some level (I don't know exactly how much) of success.. I have the problem that I am quite risk-averse and I do not think I could actually be a successful entrepreneur no matter how much I try. I have 2 homes and other valued possessions (cars, motorbike, aircraft) and a pretty comfortable life style... I have a partner and 2 children to support. Would I risk losing these things for some scheme? No, probably not. Thinking about it I actually wonder if entrepreneurs are simply illogical people because logical people would normally be expected to have assessed the risk and run the other way! It would be interesting to know what Myers-Briggs type the entrepreneurs typically are... I'm an INTP and I bet they're not! |
| Sat 27 Dec | no name | "History is rarely made by reasonable men." |
| Sat 27 Dec | Bored Bystander | [quote] I think he is taking a very well deserved poke at Joel and others like Joel whose primary marketing mechanism for their products is being an industry pundit, [/quote] I think there is a co-dependency rampant among some people wherein we look to a pundit to give us the path to success. That's the 'dark side' of blogging and web mentorship by people like Joel, Eric, Janet Ruhl, and others. Some of us waste a lot of time reading other people's opinions about how we're supposed to succeed. I snapped out of it a few years ago when I realized that glib, haughty, or one sided pronouncements of the 'correct' way to do things was holding me back much more than my own resources, talent and energy level. Reading blogs about this stuff can be a distraction, not useful information, since the proof is in making the attempt and not inferring that it's 'no use' to even try. I almost think it would be better at times to have no guideposts and mentors, and just try *something*. Just once, I'd like to read one of these pundits saying 'get the hell off of this column and do something really stupid and daring.' OK, 'fire and motion' was covered already by you know who... ;-) |
| Sun 28 Dec | Tayssir John Gabbour | I wouldn't hold this against Eric. He clearly was in this for the aesthetics, but the rational side eventually overrode it. As for the Joel .ne. Eric issue, these are two different people. What is the point in imposing an ordering on them? Incidentally I'm impressed Microsoft is hosting this. Reminds me of Gabriel's _Patterns of Software_: http://www.dreamsongs.com/NewFiles/PatternsOfSoftware.pdf 'Academics have an interesting view of business: They equate business with war. A company wins because other companies lose -- a classic misunderstanding of evolution. A company keeps everything a secret. A company tries to spy on other companies. A company cannot trust anyone not part of themselves. Because of these beliefs, AI companies rarely made strategic partnerships.' |
| Sun 28 Dec | Mark Hoffman | The 'Investing outside our field' story is a classic sucker deal that says 'I shouldn't really be in business'. And yet...here he is, complete with his Inc. 500 award... So much for common wisdom, huh? I thought it took a little bit of guts to admit to some of those mistakes. Some of those were real doozies. I certainly don't think Eric was advocating that readers rush out and do stupid things. In fact, he says as much in the article. Rather, I think he is telling people don't worry if your thought-out plans turn in to disaster. Learn from it and move on. My first business tanked in a most spectacular way. I made a lot of mistakes, but I learned from them. I picked myself up and moved on and tried again. And even if I manage to plow this one into the ground, I'll figure out what I did wrong, correct it and try again. |
| Sun 28 Dec | Charlotte C. | Gwyn, ENTP is a common entrepreneur profile. INTP and ENTP are worlds apart, remarkably. |
| Sun 28 Dec | Kerry Heffermen | Of course something like 60% of the Inc 500 firms were started by people that didn't have any experience in the field prior to starting the business. I also believe you will find all of the MB Types in entrepreneurship positions. You have to be self aware and know what your style is, and be able to sell. Introverts can be successful in sales, they just have to have the right environment. Ditto what Bored Bystander said, too. |
| Sun 28 Dec | Jim | > I would really like to know where he got all that money to make all those mistakes. He's an ex-Microsoft employee, like others we know. So I would guess that stock options is the most likely answer. |
| Sun 28 Dec | no name | AFAIK: He worked for some company that developed the early version of internet explorer. That company got bought by microsoft. He probably had a significant stake in the company that got bought. |
| Sun 28 Dec | Charlotte C. | 'I also believe you will find all of the MB Types in entrepreneurship positions.' Did I say otherwise? There are people of every personality type in every career. Yet some types are more heavily represented in some careers than others. Of all 16 types, ENTPs, according to actual studies of actual real people, are more represented among entreprenueurs than any other type. http://www.careertypes.com/entrepreneur.asp http://www.geocities.com/lifexplore/entp.htm |
| Sun 28 Dec | Kerry Heffermen | Did I say you said otherwise?
What I am saying is that if you want to start a business your Myers Brigg type or Enneagram or |
| More Linux Developers than Windows ones? | Sat 27 Dec | Shlomi Fish |
| According to: http://technologyreports.net/enterpriseinnovator/?articleID=1849 <<< Linux is becoming the development platform of choice; for example, it is expected that more developers will be developing on Linux than Windows by 2004; >>> Well, 2004 is a few days ahead, but maybe the article meant that it will happen some time into 2004. I just want to speculate what it means. First of all, it doesnt naturally means more users. The fraction of developers that are working on shrinkwrap software that many people download or buy and use is very small. We can expect Windows to have more users for the forseeable future. What it does mean, however, is that Linux is more and more becoming a platform of choice for developing applications that the majority of developers need: embedded software, web-sites software, server applications, software for internal use, computation software, academic software, developed-for-fun software, etc. Development on Windows has a higher initial cost, requires working against a proprietary environment with more flaws than Linux or UNIX has, and its vendor does not attempt to make it compatible with all other systems. (which restricts development there to one platform). So, its a partial victory for Linux, but still not the full monty of conquering the desktop. |
| Sat 27 Dec | Charlotte C. | I agree. Regarding platforms favored by various sorts of developers: 1 All free software will be developed on Linux in the west. 2 All for-profit software will be developed on Windows in the east. And for the home user on a budget: 1 Linux will be the platform of choice for those who do not pirate. 2 Windows will be the platform of choice for those who do pirate. |
| Sat 27 Dec | John | If Linux becomes dominant, this is bad for programmers, because the results of our work will get to be cheaper and cheaper. If cars become free, and everyone knows they are free, then fewer and fewer people will want to pay for accessories. |
| Sat 27 Dec | Shlomi Fish | John: I don't necessarily agree with you here. First of all, I don't understand the car analogy. Do you mean that the entire cars are made free (with all their accessories), or just the cars themselves? If just the cars themselves, then more people will buy car accessories, because they are the cars' complementary product: http://www.joelonsoftware.com/articles/StrategyLetterV.html Now as for Linux and whether it is bad for programmers: I don't think so. Granted, more software that can now be bought will be made available for free. (as in beer and as in speech). However, the number of developers who work on developing for-profit software is very small. A great deal of them (like the authors of most shareware programs) don't expect to make a very good living of it, and are doing it mainly for fun, and for the social experience. Linux makes it much easier to develop other types of software (which I enumerated in the body of the post). We can expect that if a system where such solutions are possible becomes more accessible (i.e: Linux is available and popular), then demand for programmers doing something with this system will actually increase. So, I think it would be good for programmers. (at least most of them) This is without getting into software customization and other possibilities that are better done with open-source software. |
| Sat 27 Dec | John | Shlomi, I think that the theory about complements is true in most cases. It's true in general, but it's not true in all cases. I think that when CEOs and business people will see that Linux, Apache, PHP and AbiWord are free, and that they are good alternatives, they will get the impression that software should be free or should have very little value. So, it is only logical that business people and CEOs will offer less money to programmers. They will think 'Hey, Apache, Linux and PHP are excellent and are free. Your work is not better or larger than Linux, so I don't think I'll have to pay you a lot'. So - programmer salaries WILL drop if Linux, Apache, PHP, etc will become dominant in the business world. |
| Sat 27 Dec | Tom Hathaway | 'Your work is not better or larger than Linux, so I don't think I'll have to pay you a lot'. Huh? One has nothing to do with the other. Custom developed or purchased software is a cost of doing business, like the electric bill or a truck. A business will buy it when the return on investment justifies the cost. Any sensible businessman will pay for the best price/performance of the alternatives available, and any sensible programmer will get as much renumeration as he can. |
| Sat 27 Dec | Bongo Drums | You realise you've just proved the point that Linux is stupid for programmers? You say that most programmers 'don't expect to make a very good living out of it.' At to whether there are more Linux developers than Windows ones, I think the figures for 'Linux developers' would include a lot more amateurs and wanna-be's than the figures for Windows developers. |
| Sat 27 Dec | Philo | Tom, in all sincerity, how much contract development experience do you have? Because your first paragraph tells me you haven't done a lot. Convincing people who don't write code the difficulty and value of writing code is one of the hardest things in this job. Philo |
| Sat 27 Dec | Yummy | Just because there are more people developing on Linux doesn't mean that they're doing it for free. I was part of a small consulting house that did custom server software. We'd do everything on Linux, FreeBSD, etc.. and recommend the appropriate OS for the production system (generally it was Linux or Solaris/Intel for the best price/performance). And we sure charged them through the nose for the privilege of hiring us .. usually $2-3K USD per day per developer. |
| Sat 27 Dec | John | I am sure that there are some developers getting a lot of money developing for Linux. However, I am more interested in averages. Some big questions are: 1. If Linux succeeds in the corporate world, will the average developer salary increase, or decrease? 2. If Linux succeeds in the corporate world, will there be more, or less jobs for developers? I belive that if Linux succeeds in the corporate world, the average salary for developers will decrease. Why? Because the people running businesses (who are usually business people with very little IT knowledge) will get used to the idea that lots of software is free - so a programmer's work can't be worth very much. |
| Sat 27 Dec | Tom Hathaway | Philo - Not much contracting, but I think the situation is the same everywhere. The company I work for now would never consider letting a programmer negotiate with a customer; that's the salesmen's job and they are good at it or they are gone. That's what we're really talking about here: salesmanship and negotiating. Very few programmers can sell themselves or their project. It would be interesting to hear your experiences now that you're on that side of the table, although I'll bet when the serious talk starts the engineers are out of the room. |
| Sat 27 Dec | Brad Wilson | How many of those "Linux" developers are really Java developers using a commoditized OS? |
| Sat 27 Dec | Philo | [nod] to Brad - I was just wondering the same thing. Does PHP count as 'Linux development'? If it does, then did they also count all the ASP web wannabes? The more I think about it, the less I think that there can be any value in trying to measure 'linux developers vs. windows developers' Philo |
| Sat 27 Dec | Brad Wilson | Ah, I see now the keyword which comfirms my suspicion. 'more developers will be developing on Linux than Windows by 2004' Note that says ON Linux, not FOR Linux. I don't think it's impossible. There are an awful lot of Java developers, and I imagine the great majority of them are using Linux as their day to day development environment. |
| Sat 27 Dec | Vince | Brad, very good observation. I'd venture to say a *lot*. Also, John, I don't think your correct at all. THe more companies that start using free software like apache, mysql, jboss, etc. the more companies will NEED programmers to program for apache, mysql, jboss etc. Make sense? The more companies that NEED apache programmers, the more DEMAND there will be for apache programmers. How is that not a good thing for developers like us? |
| Sat 27 Dec | John | A company wants to get an IT task done. If they use a free solution (such as GNU C++, for example), some programmers will get paid. The business people get the idea that, since GNU C++ is free, then the work of the programmers in general is not very valuable. If they use a proprietary solution, the same number of programmers will get paid. Plus, other programmers will get paid to develop the proprietary solution. Another important thing: If companies and individuals use commercial software, then a programmer or a group of programmers have the chance to create their own software company, because they can write some software and sell it. This is MUCH harder in a 'open source everywhere' world. Yes, there are some exceptions.. but if you look at the profits of software companies that sell commercial software, vs. the profits of software companies based on open source solutions, you will see that the former are doing a lot better than the latter. Don't you want to at least have a chance of starting your own software company? |
| Sat 27 Dec | Vince | and for what its worth, I love linux for server envoirments, but I wouldn't ever THINK of using linux for my workstation. |
| Sat 27 Dec | T. Norman | The increased spread of Linux and of-the-shelf software is likely to make corporations less willing to pay for off-the-shelf software. They will seek out free alternatives more often. However, most programmers aren't employed to write off-the-shelf type of software; most are paid to create and maintain customized software for a particular company. Companies aren't stupid enough to think that because they can buy a piece of tax software for $50 it means they can pay you peanuts to write programs that do tax calculations unique to their business. As it is, they are already paying us as little as they can get away with. Obtaining free software means development projects will have a greater ROI and be more likely to be approved. Stop crying that you won't be able to make money writing yet another operating system or yet another web server. Go and develop something to solve a business or scientific problem, or figure out something else to create that hasn't been already done 100 times. |
| Sat 27 Dec | T. Norman | >'If companies and individuals use commercial software, then a programmer or a group of programmers have the chance to create their own software company, because they can write some software and sell it.' If individuals didn't have free software tools and platforms, like Perl, C++, Linux, Python, Apache, etc., it would be more difficult for them to start their own company because of the cost of proprietary licenses. |
| Sat 27 Dec | Vince | Since this is turning into a discussion of how open source will affect us, let me recommend that everyone read a great article by Marc Fleury. Basically he explains why he CAN and IS making money by providing an open source project such as jboss. No small time ISV is going to be able to come up with a complex Application server (or database program, or operating system, or ...) and make money doing it. BUT, many ISV's can and will make money writing custom software FOR those big, complex applications. Hence the reason OPEN SOURCE WILL MAKE US ALL MONEY. http://www.jboss.org/modules/html/white3.pdf |
| Sat 27 Dec | Bongo Drums | Vince, developers have been writing complex applications for some time. JBoss didn't create that market and probably hasn't changed anything. T Norman, the cost of a piece of software is typically tiny compared to the cost of developers so I think the argument that it's important is irrelevant. Which leads to the bigger question of why someone would promote such an agenda. I think the argument that free software undermines the generally valuation of programming services is probably closest to the truth. |
| Sun 28 Dec | T. Norman | The combined cost of a set of proprietary software is NOT tiny for small new companies that don't know how they are going to meet next month's payroll, or for people thinking of growing a personal project into a profitable operation. |
| Sun 28 Dec | crapix | That 'analysis' is patently ridiculous. I doubt Linux would come even close to one tenth of Windows development. Furthermore, probably 90% of Linux development is php, in which the programming is done on Windows and the target browser is Windows/IE. 1/10th of 1% sounds about right. When the hype dies away in a few years, I doubt anyone but a handful of core diehards will even remember Linux. |
| Sun 28 Dec | Jim Battin | >When the hype dies away in a few years, I doubt anyone but a handful of core diehards will even remember Linux. Much like 'Linux will take over the world' this has been heard over and over since about 1995. The simple reality is that Windows, Linux, Closed and Open Source is here to stay, at least for the next 5 years. Anyway, if you're seriously worried that Open Source projects will destroy the commercial software market, you should focus more attention towards Commercial Software- as it's entirely that market's fault if OSS takes over. No offense, but if you can't do better then a bunch of 'rag-tag amateur programmers' (these forum's words, not mine), your company doesn't deserve to be in business, nor do you deserve to collect any type of salary. Professional salaries demand professional programmers producing professional software; anything less is unacceptable. Why do you think people use FogBugz instead of Bugzilla? Obviously because of the additional value FogBugz offers, it more then justifies its price and pays for itself in the long run. |
| Sun 28 Dec | foobarista | Outside the EU and US, the competition is between Linux and pirated Windows, with Linux gaining ground quickly. When I visited China recently, it was amazing just how many people there regarded Windows as yesterday's news. I asked a couple of friends about it and it basically boiled down to a combination of not wanting to feel guilty about pirating software, the high cost of MSFT products, and a fair bit of anti-Americanism - Linux is seen as much less 'American' than Windows. (disclaimer: I'm primarily a Linux or Posix-environment embedded systems developer, who develops primarily on Linux, although I do have to support stuff that runs on CE.) |
| Sun 28 Dec | Dennis Forbes | Personally I'd wager that the situation of the average IT employee would improve/stabilize if Linux made a concerted effort to invade the server room. The reason, of course, is economics -- For the majority of organizations IT is a cost-center and a completely non-core operation (which is why most see it as so 'easy' to outsource it to consulting firms) -- the budgets for IT are directly related to the revenue of the core business rather than some artificial manufactured IT requirement list. Given this, removing the 'Microsoft tax', as it is commonly termed, is more money in the pot for the rest of the IT department. Indeed, even if the TCO of Linux in a large enterprise isn't the panacea that it is often made out to be, the spendings on the Linux side will be in local organizational IT workers, rather than in filling the coffers in Redmond. That's one possible scenario. |
| Sun 28 Dec | Mike | Some random comments to the thread 'They will think 'Hey, Apache, Linux and PHP are excellent and are free. Your work is not better or larger than Linux, so I don't think I'll have to pay you a lot.' Not a logical conclusion '1. If Linux succeeds in the corporate world, will the average developer salary increase, or decrease?' If Linux succeeds in the corporate world, will the average developer shower? 'How many of those 'Linux' developers are really Java developers using a commoditized OS?' Great question. 'Much like 'Linux will take over the world' this has been heard over and over since about 1995. The simple reality is that Windows, Linux, Closed and Open Source is here to stay, at least for the next 5 years. ' *nix is like a star that is going to supernova. Linux is the supernova, after that there sure is a lot less star (*nix) Remember *nix is 30 year old foundations. There are fans (mostly sys admins), but the people that choose it at the corporate level would probably rather buy Windows, but are forced to *nix because they NEED something Windows yet lacks (security, scalability, etc) Once Windows comes on 72 processor boxes and is truly a server os, not one line with all the home user features there, Unix will fade. Anyone that mentions the word Dell and scalability in the same sentence does not fathom the type of scalability I am talking about. |
| Sun 28 Dec | Dennis Forbes | 'There are fans (mostly sys admins), but the people that choose it at the corporate level would probably rather buy Windows, but are forced to *nix because they NEED something Windows yet lacks (security, scalability, etc) Once Windows comes on 72 processor boxes and is truly a server os, not one line with all the home user features there, Unix will fade.' Windows Server 2003 (aka Server XP) has a 64-bit edition running on massive Itanium boxes - An 64-cpu Itanium2 HP Superdome running. Without talking about rare research computers, this is generally about as powerful scaled up as systems get in the real-world enterprise space. Having said that, Linux isn't taking over on the big iron -- it's taking over on the small server systems, which comprise the vast bulk of computing center inventory - little one, two, maybe four way servers serving various group apps, etc. People often choose Linux not because of the features it has that Windows lacks, but rather because it has the features that they need for their project to succeed ... why pay for Active Directory or any of the myriad of other Windows features if they are totally inconsequential (and indeed of negative value) to the project? |
| Sun 28 Dec | Charlotte C. | 'However, most programmers aren't employed to write off-the-shelf type of software; most are paid to create and maintain customized software for a particular company.' I disagree - there are more app developers than there are custom one-off database setter-uppers. Not ever sure why the data base administration set up tasks we are talking about are even being considered in the same caliber as real programming. Now I do agree there are more database administrators than app developers, and also more network administrators than app developers and so these are a couple of smart fields to go into since they are less likely to be outsourced. And also a lot less skill is needed to throw together some queries and forms than to do real world app development that has to be usable by Aunt Em and powerful to Geek Gary. But when it comes to actually real programming, most developers work an mass marketed software, not one-off software. Very few companies develop custom apps from scratch nowadays. But lots of people buy new games and other mass-market programs. |
| Sun 28 Dec | Philo | 'They will think 'Hey, Apache, Linux and PHP are excellent and are free. Your work is not better or larger than Linux, so I don't think I'll have to pay you a lot.' Not a logical conclusion ************** No, it's not. Never worked with corporate management, have you? They are not exempt from irrational thought. Philo |
| Sun 28 Dec | Shlomi Fish | Well, it seems that this thread sparked a very active discussion overnight, which is good, I suppose. Here are more of my thoughts. I think we can safely assume that developers who write software on Linux would usually intend it on Linux. While there's Java, Perl/Python/Tcl/whatever, wxWindows, and other cross-platform solutions, you cannot really guarantee they will run correctly on Windows, unless you've tested them there. Some developers still develop for Linux while sitting on a Windows workstation (so they'll have MS Office, etc.). But generally, Windows and Linux/UNIX are so radically different, that most developers will have to choose one over the other. So, we can assume most development will be intended for Linux (or at least for both UNIX and Windows) as well. I again stress the fact that programmers will probably not get paid less to develop on open source platforms. Again, if a company wishes to get its job done, it will want to hire a clueful developer. Even if it does not require any specialized software, it will still need a clueful crew to maintain the computing facilities. (and in the UNIX world a Sys Admin is usually a programmer on ice) The wages of programmers are determined by market dynamics, by how much the organization is willing to pay for them, etc. rather than whether they work on free software or not. A great deal of programmers worked in LAMP environments, (UNIX, Apache, MySQL/PostgreSQL, and Perl/PHP/Python) and still got adequate wages for years. As for whether proprietary software will still exist in an open-source-dominated world (assuming such a situation will materialize) - I think it will. Some proprietary software is of exceptional quality and scope that no open-source solution of its quality, is yet available . This situation may change, eventually, for individual packages, but as a whole there is a place for proprietary enterpreneurship and innovation. Also, even in markets where there is adequate bug tracking software, one can many times make money selling proprietary software. Joel is still selling FogBUGZ for money, despite the fact that bugzilla, Mantis and many other open source packages for tracking bugs are available and commonly used. (not to mention a lot of other proprietary competition). And naturally, the majority of developers don't work on any kind of shrinkwrap software (open source, proprietary, consultingware, etc), so most of them are generally not going to be out of jobs. |
| Sun 28 Dec | NDA Rules | I've actually encountered the open source effect. An 'open source' web firm commissioned me to develop a sophisticated J2EE backend to perform critical processing for a web site for a government client. When I told them how long it would take and what it would cost, they were incredulous, literally pointing out that they could get this and that software for free, and that surely adding a few little bits couldn't possibly take more than a few days. They didn't understand that the rubbishy little bits of PHP they showed me were worthless, and would form no part of a well designed Java project. The thing that got me was that 1) they seemed to have won the project by offering to 'use open source solutions,' when they weren't actually capable and 2) while they quibbled about my fee, they had no qualms in billing the government for all of their own time. |
| Sun 28 Dec | Stephen Jones | Dear NDA rules, This is not the open source effect at all. If you were doing the project on MS SQL server they would still be incredulous that for a simple VB.Net or VBA solution you were charging them tens or hundreds of times more than they paid for all the powerful software they got from Microsoft. Nor is your problem limited to programming. Any lawyer, corporate trainer, private tutor, plumber or electrician will tell you that people are notoriously averse to paying the going rate for any service. |
| Sun 28 Dec | Shlomi Fish | Re: Linux in China. Well, I don't know if this trend is significant there or not. (I did hear a statistics about Thailand that indicated this was the case). The 'anti-American' opinions they have expressed probably has some anti-Capitalistic roots. I, (and Eric S. Raymond) would very much disagree that Linux is anti-Capitalistic, but I can understand where the analogy is coming from. Even in countries where Capitalism is prelevant, people are inherently not very willing to pay a lot of money for software that can otherwise be distributed for free. I was told, that in Economics, a vendor who sells for money a product which had a certain development cost (possibly very large) and its production cost is zero is called a parasite. (!) But I suppose that whatever trends exist to use Linux, whether in developing or developed countries are positive, because I find it a great product regardless of price. |
| Sun 28 Dec | NDA Rules | Stephen, you're wrong. I've been in business a long time and delivered good work to lots of companies big and small and they usually know what they need to pay and are happy to pay it. This was quite explicitly an effect where they wrongly extrapolated from the availability of free source code to the cost they expected to pay for sophisticated software development. |
| Sun 28 Dec | Mike | 'Windows Server 2003 (aka Server XP) has a 64-bit edition running on massive Itanium boxes - An 64-cpu Itanium2 HP Superdome running.' Itanium is dead. I should have added 72+ procs on a chip architecture with a future. Being a sysadmin type myself, I prefer Unix. I like Linux over Windows and Unix over Linux. I bash the Linux zealots a lot though becasue sometimes Microsoft solutions do make sense. |
| IDE RAID | Sat 27 Dec | Philo |
| Two questions... 1) Im seeing a lot of higher-end consumer PCs coming with RAID 0 installed by default. Has anyone bought one of these? Im wondering if they come with the huge warnings *I* would put about the dangers of striping (namely youre cutting your MTBF by the number of drives). 2) Does anyone know if there are any IDE RAID 5 solutions? I want to increase my storage at home, and Im currently leaning towards a mirrored pair of 160 GB IDE drives, but wouldnt mind going with four 40GB in a RAID 5 array (I know, 25% less storage). ...and Ive gotta go do some SCSI math again... Philo |
| Sat 27 Dec | John Rose | The IDE RAID-0 solutions seem to be mostly aimed at the PC-hot-rodding 'enthusiast' market. I don't think they know or care about the obvious data risks, they just want every bit of performance. Luckily all the IDE RAID solutions seem to do RAID-1 as well, for a more sensible approach. I'm sure IDE RAID wouldn't be very popular with only RAID-1, so I'm glad all the RAID-0 knuckleheads are subsidizing my cheap RAID-1 setups. :P Not sure about IDE RAID-5. Never heard of any manufacturers implementing it, that's for sure. |
| Sat 27 Dec | Philo | Yeah, hard drives are cheap enough to make RAID-1 fine. I actually got to the point of thinking of rebuilding my machine from the inside out, but I think I'll stick with upgrading CPU, video and hard drives, then looking at a new box later in the year. Philo |
| Sat 27 Dec | Jim | Philo, These 3 articles might be quite useful for you: http://www6.tomshardware.com/storage/20021112/index.html http://www6.tomshardware.com/storage/20020813/index.html http://www6.tomshardware.com/storage/20020830/index.html |
| Sat 27 Dec | Jim | Oh, and this recent one too: http://www6.tomshardware.com/storage/20031128/index.html This solution gives RAID-5 SCSI speed using IDE, but at a cost of only about 10% higher than standard IDE RAID solutions. Check it out. |
| Sat 27 Dec | coresi | High end consumer motherboards come this days with RAID0 (stripping, basically if you lose 1 HD all data is lost), RAID1 (mirroring, after losing 1 HD the information is still intact) and RAID0+1 (nicely described as mirrored pair of striped arrays). I am leaning towards a 120GB RAID1 myself. There are 2 reasons for choosing a RAID: fail-safe data and performance. RAID0 is not fail-safe while RAID1 has a lower performance due to the write penalty. RAID5 is a more expensive combination to alleviate those problems. Anybody has experience with four 7200 RPM hard drives in the same box? I think they will need 2 cooling fans only for themselves. |
| Sat 27 Dec | Philo | I've got four older 7200rpm SCSI drives in a box with two P3's - my box with a single P4 and one drive runs hotter. Philo |
| Sat 27 Dec | Dan G | http://www.3ware.com for all your IDE RAID needs Unlike most other RAID controllers and most of the ones that are inbuilt into the motherboard, the 3ware cards do the majority of the RAID processing in hardware, whereas others will take a slice of CPU time |
| Sat 27 Dec | Brad Wilson | Dan makes an important point. These $10 RAID chips on the motherboard rely almost entirely on software to perform their RAID functionality. The system cannot be ignorant of their presence (as it can in the case of a fully hardware RAID) because the drivers are doing the heavy lifting. RAID 5 is definitely alive and well in SCSI, but the cost is hard to justify for anything but servers. I'm curious about the RAID 5 IDE link that he just posted... I run a SATA RAID 1 120GB array for my desktop development, and it's fast. Not as fast as a fully hardware solution would be, but it's certainly never felt in the way (like a fully software RAID 1 solution does). |
| Sun 28 Dec | John Rose | Thanks for the link to those 3Ware RAID cards. they look quite nice. And in relation, they do support IDE RAID-5, by the way. For the record, I've been using an el-cheapo Highpoint IDE RAID card on my development server for a RAID-1 config. I'm sure it's not as fast as a more hardware-based RAID setup like the 3ware cards, but for what it's worth the Highpoint card+drivers have been an absolute breeze to set up and use. Nothing but smooth sailing... |
| Sun 28 Dec | www.marktaw.com | My MoBo came with RAID but I never used it. The RAID channels though were in addition to the existing IDE channels, so I think my MoBo can support up to a total of 8 drives, or 6 if I want to use RAID. The same cautions as everyone else about RAID 0... I know someone who just lost all his data, I presume it was because of RAID issues. |
| Sun 28 Dec | anon | Adaptec make an IDE RAID-5 card, and very good it is too. It supports upto 128Mb cache in the form of an 168 pin SDRAM DIMM. I would strongly recommend the cache be invested in. Running at RAID-5 on IDE means there can only be one drive present on each channel, on the Adaptec cards this is true at least. Something to do with IDE not being able to read/write to multiple devices on the same channel simultaenously I believe. |
| Software cloning and copyright | Sat 27 Dec | The Thinker |
| Ive seen lots of software being cloned. I am speaking mainly of simple software not something like Photoshop or Dreamweaver. But anyway, is this legal? And lets suppose someone clones the app absolutely perfectly, with the menu positions and captions and message box texts and whatever. Will this still be legal? |
| Sat 27 Dec | Michael H. Pryor (fogcreek) | You'd be better off asking a copyright and patent lawyer, but my instinct is that this is perfectly legal. See this for more info: http://www.uspto.gov/web/offices/pac/doc/general/whatis.htm The things you are dealing with are patents, trademarks and copyright. Patents protect you from having others make your widget.. this can be a program. Search google for 'software patents' for all the latest hoopla surrounding this. Provided you have no patent on your program, then is any piece of the program trademarked? Such as a logo or design of the interface? (Maybe a Nike program has its window in the shape of a 'swoosh'. You obviously couldn't reproduce this without violating their trademark.) Finally, the copyright... did you copy any of the source code which was copywritten? You would probably get in trouble if you copied the text verbatim from the captions which is also copywritten. A year or two ago, someone decided they would take our entire FogBUGZ 1.0 program and reproduce it exactly... word for word. Basically as long as they just reproduced the design and layout of the program, we couldn't stop them. This smartie copied our entire help pages and all of our marketing material too (except he did a search and replace on FogBUGZ for his 'new' program name). As annoying as this was, we simply informed him that the text he was using in his program and website was a copyright infringement. He proceeded to replace all the text in the program and his website with really bad english which was a hoot to read. We figured the rewritten text was enough of a bad omen for people to realize this guy was shady. 'Imitation is the sincerest form of flattery' (Colton, Charles Caleb (1780-1832)) |
| Sat 27 Dec | Joel Spolsky | Oh and read up on all the old Borland/Lotus "Look and Feel" lawsuits. |
| Sun 28 Dec | Matthew Lock | This makes interesting reading 'Against User Interface Copyright' http://lpf.ai.mit.edu/Copyright/look-and-feel.html |
| Selling software applications to Microsoft | Sat 27 Dec | Enrique A. Woll B. |
| This is probably pure fantasy, but I have been wondering if it is possible to sell simple applications that run on the Office platform to Microsoft. The idea is that these would be packaged within Offices Excel as powerful working examples of what can be done with the spreadsheet program, and be sold by the millions each year even if for pennies per copy. Email encryption, weight-related clinical advise, and real estate rent design might be appealing application areas, amongst hundreds of others. I realize that there is nothing new about the idea, which in a way parallels what Bill did with DOS and IBM, and that there are literally millions of spreadsheets out there, but please wake me up someone. Thanks! |
| Sat 27 Dec | Michael H. Pryor (fogcreek) | The problem as I see it, would be that it would be highly unlikely for Microsoft to invest the time necessary to broker a deal with you for some working examples, when they could pay a college intern to write the same. This isn't to say that you're working examples wouldn't be useful or worth money, but only that Microsoft wouldn't bother to enter into a revenue sharing deal with you since the value to them would be negligible. The amount of money they would spend on their lawyers fees just to email you a few times would already dwarf the revenue they could make selling these things. You'd be better off writing them yourself and setting up as an ISV. Give away sample ones, and charge for the full versions. Do a google search for 'office add-ins' and you'll see there are a bunch of websites already devoted to this. |
| Sat 27 Dec | George | GeCAD, a Romanian company, has managed to sell their antivirus product to Microsoft. http://www.ravantivirus.com/pages/shownews.php?i=153 You could search for more information about how they did it. |
| Sat 27 Dec | Enrique A. Woll B. | Thanks Michael and thanks George! Enrique A. Woll B. |
| Sat 27 Dec | Mitch & Murray (from downtown) | We've sold some of our low-level development tools to Microsoft years ago, and all I can tell you is (1) I don't know what they did with them and (2) they paid their bill on time. I don't think selling to MS is the path to riches unless you have their ass over a barrel in some sort of patent dispute. |
| Sat 27 Dec | Joel Spolsky | Many of the add-in tools in Excel came from outside vendors. For many years the Excel 'solver' was a low-end package from an outside company that also made a high-end package. |
| Sun 28 Dec | Philo | Incidentally, *that* is the path you want- the flexgrid/crystal path. Philo |
| Sun 28 Dec | Enrique A. Woll B. | Thank you Mitch & Murray! I guess you have read the preceding comments, and I wonder what you would think of the Non-Disclosure Agreement proposed by GECAD Software of Romania for use with contributing software developent entrepreneurs that wish to team up with them in the conceptualization, development, and marketing process for new software tools and applications? The contract is brief, and can be read at: http://startit.gecad.ro/start/ Thanks and regards, Enrique A. Woll B. |
| Sun 28 Dec | Enrique A. Woll B. | Thank you Joel Spolsky! I never fail to be amazed at how Excel, a product that clearly bested the competition in practically all respects when it was released, remains relatively dull, with true creativity for any significant improvement of its cutting edges seemingly coming consistently from outside MS! In my case, what I propose to sell is not necessarily add-ons, though I have developed what could become very powerful functios for complex cash flows and related. Rather, I propose for the time being, to market full applications that are not run-of-the-mill and appeal to wide lay audiences, and that run on Excel. Enrique A. Woll B. |
| Sun 28 Dec | Enrique A. Woll B. | Thanks Philo! I appreciate your suggestion. Enrique A. Woll B. |
| Recovering data after HD hosed by Magistr virus? | Sat 27 Dec | Damn them virii |
| Hi Although protected by McAfee (and ZoneAlarm), my mums computer was hosed by the unfamous Magistr virus (all her Eudora mail files removed, most of her .DOC files removed with some of them remaning by with the YOUARESHIT string prepented.) More infos here: http://www.fabian.com.mt/VIRUSalerts/VIRUSALERT57.html I did recover the drive using the open-source TestDisk applet, but obviously, the files are gone. Since the drive also had its MBR hosed, my mum could not longer boot... which is a blessing because most likely, the data are still there. Anybody knows of a good utility that could try to recover the sectors? I heard of an app called Tiramisu, but never used it myself. Thx a bunch |
| Sat 27 Dec | Damn the manufacturer | Merry Christmas, From Microsoft |
| Sat 27 Dec | Damn them virii | ;-) |
| Sat 27 Dec | Insert half smiley here. | Sorry, I can't stop myself: the plural of 'virus' is 'viruses'. 'virii' would be the plural of 'virius'. google gave me this link about it: http://www.perl.com/language/misc/virus.html I use a hex editor called 'frhed'. It has a 'Open disk' option on its menu, with 'goto next sector' and so on. (I hope you'll forgive my not trying it on my PC!) frhed has a search capabitility, which I presume works with disks too. So, this may do what you want. (If the files are fragmented, you are probably out of luck, but you never know.) frhed is here: http://www.kibria.de/frhed.html Good luck! |
| Sat 27 Dec | Damn them virusES | Thanks, but you can imagine the amount of time and the difficulty it would take to rebuild files manually that spread over multiple sectors... which is just about any file since sectors are usually 512 bytes... ... provided the virus doesn't corrupt files before deleting them, which Magist does. I recovered a bunch of files using Ontrack EasyRecovery (previously Tiramisu)... only to find that their contents had been corrupted before deletion. I hear AVG is a good antivirus ;-) Thx anyway |
| Sun 28 Dec | Mike | 'I hear AVG is a good antivirus ;-)' My mom has run this for 2 years without incident. |
| Sun 28 Dec | Damn them virusES | I guess the McAfee antivirus that came with her HP Pavilion is a good program... but more so if its auto-update feature is enabled ;-) I'll write some batch file with NcFTP to automate backing up her important files to her ISP. |
| Statistics - reading code vs. writing code | Fri 26 Dec | JT |
| Im looking for some quotable statistics for some marketing materials. I highly suspect that many developers spend more time reading code (looking for bugs, or just plaing maintenance) then writing new code. Does anyone know where I might find some quotable statistics about this? ...or to the contrary, if Im incorrect. Something like an academic paper or other reference-able source would be great. And for that matter, what about yourselves? Do we spend more time reading existing code, or writing new code? |
| Fri 26 Dec | sgf | I just finished reading Code Complete by Steve McConnell and I'm pretty sure ha gave some statistics in there. Unfortunately I don't have it here right now... |
| Fri 26 Dec | Tom Vu | 'And for that matter, what about yourselves? Do we spend more time reading existing code, or writing new code?' Does a writer spend more time writing or reading? i think it's obvious most programmers spend more time reading than writing. |
| Fri 26 Dec | MSHack | Not to sound like a famous politician, but it depends on what we mean by 'read.' consider any single line of code: x=y; I wrote it once and at a minimum will read it twice. Once while actually typing it and again to verify completeness. Consider any more complex structure: // statements_for.cs // for loop using System; public class ForLoopTest { public static void Main() { for (int i = 1; i <= 5; i++) Console.WriteLine(i); } } While I consider myself fairly good, I would be kidding to say I could type even this simple example, error free without reading it, nor would I want to. |
| Fri 26 Dec | Ori Berger | The *time* spent reading while writing is, I think, attributable entirely to writing. If you insist, adopt a simple rule that says that, on a first day a line of code comes into existence or is modified, you attribute the time it took entirely to writing; From that moment on, every time you read it, it's reading. I think that even for productive programmers, reading:writing is 90:10 or even 99:1. And before you claim that sounds ridiculous, consider that the industry average of production quality lines-per-hour produced (meaning, reasonably bug free and spec compatible). is, last I heard, ~ 10. Now consider, where all that time is spent. |
| Fri 26 Dec | Brad Wilson | Not necessarily on reading code, though. |
| Fri 26 Dec | T. Norman | The 10 lines per day is an average for production-ready code, not including comments or whitespace. The 'days' includes the time in design, and the testing and debugging as well as other activities like meetings, brainstorming, and requirements gathering. I just finished a project that took about 4 months for design and requirements, with 1 month for coding and another month to make it production-ready. So the code produced would get averaged out over 6 months. |
| Fri 26 Dec | Dennis Atkins | Ori, where did you hear 10 per hour? Last I heard was 13 per day average and 40 for top producers. Those numbers are surely old though and I have suspected productivity is higher now and have been looking for some decent stats. Thanks. |
| Fri 26 Dec | Dennis Atkins | And yes, it's understood that these are averaged out across total effort over a year and its non-space non-comment. |
| Fri 26 Dec | T. Norman | Those numbers are also net of rewrites and deletions. If you write 500 lines, then subsequently modify 100 and remove 200 of them to make it production ready, it counts as just 300. |
| Fri 26 Dec | T. Norman | 40 lines for 'top producers'? I guess you mean 'top LOC producers' and not 'top programmers'. Because from what I've read, the top programmers don't really write much more or less code per day than average. They just get much more functionality out of their lines of code, by finding simpler solutions to complex problems. |
| Sat 27 Dec | Me | JT, there are many different types of developers. Maintenance programmers and oss people will spend most of their careers reading code. Developers will spend very little because they create new code. |
| Sat 27 Dec | Brad Wilson | I agree with T. Norman. Better developers tend to write less code, not more. They also tend to relish in ripping out code whenever possible. What's that quote? 'The sign of product maturity is when the code-base starts shrinking instead of growing.' |
| Sat 27 Dec | T. Norman | Still, in extreme cases the top developers probably do write more lines of code per day AND get more functionality per line. Otherwise Doom would probably still be unfinished if they were working at a 10-20 LOC/day rate. Project size also affects things. In general, the more people on a project is the less % of the time you spend coding. So on a smaller project your higher LOC/day can mean actual increased productivity, and not just redundant fluff. |
| Sat 27 Dec | Me | Maybe you've never worked with crap developers, T Norman. These are people who will cut and paste huge blocks of garbage, fiddle with it till it sort of does what's required, then have a great talk with the project manager. They're usually good talkers. |
| Sun 28 Dec | Pakter | I'm happy to see someone ask this question. I've asked it once during a meeting. I gave my personal answer (95:5) and they looked at me like as if I was an E.T. One of my colleagues even answered 5:95. My conclusion was that, as far as code is concerned, accelerating the reading phase is more valuable than accelerating the writing phase. |
| MySQL or Postgress | Fri 26 Dec | Mauricio Macedo |
| Hey all, Im working on a project, to bring something like Friendster to my language - portuguese. I lost a good opportunity to do this with blogs in 1999 (opendiary at the time), because I lacked the resources to keep a server up serving a million people updating it daily. Friendster isnt as resource-hog as a weblog, but it has viral marketing features that makes it grow very fast. Im worried because my current tools (ASP & Access/SQL Server) do not scale well. Im developing the website in ASP, but will convert it to PHP before launching. Now, my server supports MySQL and Postgress databases. Which one should I use? I want speedy, and will not have transactions. But I may have to change my host provider in the future, so I will also need to be able to backup/restore databases with easy. Am I wrong to think that a PHP/(MySQL or Postgress) combo is way better than ASP/(Access or SQL Server)? Ive heard of php websites with moret han 200 users per minute and working well with only a dedicated server. Is that really feasible? Thanks for any suggestions! |
| Fri 26 Dec | Mike | 'I'm worried because my current tools (ASP & Access/SQL Server) do not scale well.' Yes, I could not agree more. You need full blown J2EE for your site. There are no large successful sites written with Microsoft technologies. None. |
| Fri 26 Dec | Mauricio Macedo | Oops, make the options PostgreSQL or MySQL. Sorry! |
| Fri 26 Dec | Jim | I use PostgreSQL on Linux, it is more solid and more SQL-99 compliant than MySQL. Even though MySQL might be slighly faster in certain situations, I'm more concerned about reliability and compatibility. If you want to run on Windows, then your only choice is MySQL. |
| Fri 26 Dec | M | Mauricio - Sounds like your description lends itself to MySQL. It is fast, you don't need transactions and most web hosting companies use the PHP/MySQL combination (or so it seems from where I'm sitting). Mike - 'There are no large successful sites written with Microsoft technologies. None. ' Does dell.com not count? Seems like they are kind of successful. |
| Fri 26 Dec | Mike | I was being sarcastic to the op's view that sqlserver and asp is unscalable. My point was there are many sites that show that "I'm worried because my current tools (ASP & Access/SQL Server) do not scale well." is pure ignorance. |
| Fri 26 Dec | Mauricio Macedo | Mike - I don't a spare 1 Gb of RAM for J2EE. My website is very, very simple. I'm very surprised of the simplicity of Friendster, and how it was able to reach 1 million registered users. The /concept/ is strong. m - you're right, seems like MySQL is the best option, as they even created an acronym for the solution - LAMP! *laugh* I will reasearch a bit more about the backup/restore functionality, though. thanks for the input! |
| Fri 26 Dec | Mauricio Macedo | Mike wrote: I was being sarcastic to the op's view that sqlserver and asp is unscalable. My point was there are many sites that show that 'I'm worried because my current tools (ASP & Access/SQL Server) do not scale well.' is pure ignorance. Maybe I wasn't clear enough. They do not scale well according to my budget. I want to serve thousands of users daily, with a simple server. I will start with a shared server, and can afford a dedicated one in the future. But I can not afford a server farm. Now, if you tell me of a windows box with ASP/SQL Server capable of serving 100 different users per minute without coughing up or crashing and costing $50-$200 per month, then I'll be quite impressed and very thankful! Then again, I have heard of LAMP boxes doing just that. If this is really true, it means that I will be able to support my project for months without outside investors! Mind you, if I am getting into all the trouble of using another platform than the one I know, it is because I know the limitations of the ASP platform for high traffic websites and am trying to find a better solution. |
| Fri 26 Dec | Philo | Mauricio, don't take our word for it - you can get SQL Server on a 120-day demo for the cost of postage. Then get any of the ASP or ASP.Net demo sites available online, or write your own test pages. Then get the Web Application Stress Tool (WAST), the Microsoft Application Center Test (with Visual Studio.Net), or any other stress test tool. And bang on it yourself. I know in a previous contract I had an app with hundreds of users accessing |