| last updated:06 Aug 2003 14:43 UK time |
 |
| JOS Statistics - Recent Comments (Comments added for week ending Sun 27 Apr 2003) | View Other Weeks |
| Need book recommendations on various subjects | Sun 27 Apr | Bob Yu |
| Part of my new jobs duty is to be a system administrator (and programming in ASP using MS CMS 2001). I have never worked as an administrator so I need something that combines the fundamentals to advance optmizations. I need recommendations for Windows 2000 server administration, SQL Server 2000 administration and programming, Commerce Server 2002, and finally .NET stuff, especially ASP.NET and ADO.NET. I am looking for books that are of the same caliber as Peopleware and Code Complete (both of which I have), meaning that it is comprehensive (fundamentals to advanced), thorough and worth every dime that I spend on. With the profusion of books out there, I have no idea what is good or not, so I need your help in making a choice, for I dont have too much cash to spend on books. Thanks in advance. |
| Sun 27 Apr | JWA | Hi Bob, I've found that my favorite's for both layout and content tend to be the MS Press offerings. The Step-by-step books can be good intros if you're starting from scratch in an area, but if you get into the detailed use of any specific technology they can be somewhat basic as reference manuals. For .NET, I have and like some of the Core Reference books. They're generally rather large and thorough. --Josh |
| Sun 27 Apr | | You need a manual not Peopleware and Code Complete type books. |
| Sun 27 Apr | Philo | O'Reilly has a pocket handbook on Win2k command line utilities I'd recommend. Philo |
| Sun 27 Apr | UI Designer | Try the Microsoft website. I remember they had an online seminar about Commerce Server (or at least the Business Desk) there which provided some excellent info when I did an evaluation of CS a couple of years ago. Also look at the material on the dotNet Show. |
| Software Startup Books | Sun 27 Apr | Anthony W. |
| Could anyone recommend some good reads on starting up your own software firm? Im interested in ones that describe the problems and issues commonly encountered, and suggestions of what to do to. Really, a How To book as opposed to a touchy feely warm fuzzy one. Thanks! |
| Sun 27 Apr | | This website might help http://stylusinc.com/business/india/cultural_tips.htm |
| CNN Article on buggy software and it's cost. | Sun 27 Apr | Mark Hoffman |
| Given the recent thread on software engineering and the future of our industry, I thought this article on CNN was interesting. http://www.cnn.com/2003/TECH/ptech/04/27/buggy.software.ap/index.html Nothing really new in the article, except that its not too often you see this kind of stuff in the mainstream press unless there has been some recent catastrophic software failure. |
| Sun 27 Apr | Matthew Lock | Software is often buggy because it's developed using 'cut to the bone' budgets. Software is very expensive to develop and when a manager asks you to get it on a tight schedule testing is often the first thing to be cut from the project. |
| Sun 27 Apr | Tj | My theory is that software doesn't 'look' like anything. If you make something really small, really large, or really intricate, people respect it. But with software, it looks all free, as if suddenly a computer's resources are limitless. Most people don't live in the virtual reality, so it's understandable. |
| Sun 27 Apr | Dennis Atkins | That article is just Watts Humphrey stirring up some publicity and promoting heavy process over business realities. Regarding the dishwasher, 1. The washer was defective and the bozo who bought it should have taken the washer back to the store and demanded a refund plus a credit for installing another one to compensate him for his time lost. It's his own fault for not demanding a refund and calling a press conference instead is just silly. 2. What kind of bozo buys a computer driven internet enabled washing machine/refrigerator/toaster/blender anyway? All these mechanical devices are rugged hardy devices that work better without no stinkin computer. Regarding the Mars Lander, as I understand it the software is thought to have turned off the engines because it mistakenly thought the ground had been struck, which was detected by some sort of impact sensor that was triggered mistakenly when some other system deployed. This isn't a software bug but a design error and possibly one that may not have been found no matter how much time was spent in design, without the foreknowledge that the failure had occured. In no case would additional software testing have done anything to prevent the problem from happening and more than debugging a OS kernel will do anything to fix the problem of the CPU overheating because the case is improperly vented. I have more to rant about but I will stop now and go get something to eat. |
| Emailing resumés | Sun 27 Apr | anon |
| When I approach someone with an email, should I send a resumé with it, just link to one, or both? I think just linking to one is a bad idea since my server just might decide to go down right then, of all times. Should I not send the pretty html formatted one, and instead a plaintext/rtf? (I hope pdf isnt de rigeur.) Maybe this is a strange question; its just I get vaguely paranoid when I get sent an html attachment. Thanks. |
| Sun 27 Apr | Bored Bystander | Since nobody has jumped in... I've found that all that HR and recruiting agency droids can generally deal with is a Microsoft Word DOC file as an attachment. They're just too stupid and one dimensional to deal with anything except a file attachment, since it's what 'everyone has to do'. I once directed a recruiter to a link on my web site which had my resume, and the woman acted like I was giving her a pocket knife and telling her to remove her spleen with it. Stupid assed gatekeepers. Ok, I said stupid enough times. HOWEVER - I don't generally think that emailing a resume is a very good approach. You don't get anyone's attention that way. Companies get zillions of emailed resumes because the cost and effort of sending resumes by email is almost nothing. The best approach for finding a position is personal. Call the company, and try to at least find out HR's priorities for the position. Best case, track down the hiring managers. Try to find someone in the company who wants to see *your* resume and then send it to actual human be'ins. |
| Sun 27 Apr | anon | Good lord, I just converted my html to .doc format usinjg msword 2k, and whever I open it, it complains, 'The dimensions after resizing are too small or too large.' I don't think I can take the IT world. |
| Sun 27 Apr | note | Recruiters like word documents because it can interface with their resume database systems. |
| Dave's Forum on CAMEL | Sun 27 Apr | Philo |
| I put Dave Behnkes Forum software on CAMEL, tweaked the style sheet a little (YE GODS!). Down the road I hope to link posts on CAMEL to topics in the forum... But for now its functional, and future edits shouldnt affect content. Philo |
| Sun 27 Apr | Dave B. | OMG! Wow! This is the highest compliment anyone could pay me! You wouldn't believe how satisfying it is to know that someone is using something you wrote. ( Or maybe you guys would since you work in the business. ) Thanks Philo! ( Hope integration isn't giving you headaches! ) |
| Sooner or later dominant companies don't dominate | Sun 27 Apr | Mike |
| What will stop Microsoft from being dominant? |
| Sun 27 Apr | choppy | When Bill Gates retires or dies, the company will begin to fall apart. It's really all about Bill. |
| Sun 27 Apr | Codex | - A company which will be better at Marketing than Microsoft ? Or more simply a few Al-Qaeda terrorist who crashes a few planes on Microsoft Campuses ;-) - Do you know if they backup their source code on a remote server in Africa or something :-) |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | Actually, I disagree that it's all about Bill. I actually think it's all about Ballmer. |
| Sun 27 Apr | Philo | As I've said before - I think Ballmer is the reason they're starting to slide. Despite his monkeyboy act, he doesn't 'get' the whole developers thing. I also think he doesn't understand 'penetration, then profit' Philo |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | I can't think of a company or time that's been better to developers than today's Microsoft. Why do you feel they're doing disservice to developers? |
| Sun 27 Apr | Philo | It's actually gotten better of late, but about two years ago MS was heading down a nasty road: 1) Threats of audits on MSDN software 2) MS Passport, as initially priced, was something like $15k just for a developer to enter the market 3) Drop in academic product availability 4) Product activation on MSDN operating systems It just seemed to me for a while there that MS was turning its back on developers. Seems to have improved since the release of .Net. Philo |
| Sun 27 Apr | Bella | Actually, Ballmer said it's all about the 'Developers.....Developers.....DEVELOPERS.....' http://www.zdnet.com/anchordesk/stories/story/0,10738,2807333,00.html http://www.detonate.net/media/dancemonkeyboy.mpg |
| Sun 27 Apr | Michael Kohne | I don't presently develop for Windows (I have done so in the past), but I personally think MS has already made their mistake - it's just a question of whether it will catch up to them before they do something else for other reasons. The mistake? Trying to force everyone to a rental software model. Why is it a mistake? Because previously, if you were an IT type, you might budget 'X' for computers, then be able to decided later whether you wanted to pay more money for the latest software rev or not. If it had features that helped your buisness, you budget and buy. If not (or if you are low on money) you don't buy. Now, with a rental model you don't have the choice - you pay yearly, whether you upgrade or not. And when your vendor screws up (recent Office 2K issue), you run into trouble. Or if you are struggling and pay your bills late - all your software dies. I have the feeling that when the realities of a licensed software model really hit home, large companies are going to switch to solutions that DON'T come with this kind of risk attached. But that's just an opinon. |
| Sun 27 Apr | | I'm a me-too'er re software licenses. But, I think microsoft is smart enough to make the correct decision (whether I'm right or wrong). I trust Ballmer to make the right decisions. |
| Sun 27 Apr | Nat Ersoz | Tides change, IMO, due to some truly disruptive technology coincident with a power vacuum. In 1980, it was mainfraines quickly giving way to mini's, which gave way to the '386 in 1992 and the introduction of Win95 - the first offering from MSFT that was viable. While Apple was always providing superior solutions - in every area - during this period, it was the power vacuum that was filled by MSFT. Similarly, in 1974, with an oil crisis and gasoline quadrupling by 1979, Honda and Toyota filled the vacuum with reliable economical cars which nearly put Detroit under. So, I think it takes not only a disruptive technology, but inability of the status quo to react with sufficient speed in order to bring down the Goliath. Right now, MSFT suffers from an attack of the Lilliputians (sp?) . But they react much faster and more effectively than will be required for their fall. Certainly the day of stock prices doubling every 2 years are likely over. And they need to be oh so wary of profits coming from 2 rather narrow sectors that well could run out of steam. But... At present rates, a sea change is not immenent. On another almost orthogonal note, technical analysts are increasing bullish in general. I'm saying the next down stop for MSFT is 23, maybe 23.5. And it might not even make it that low prior to a break out above 26. Might be a really good short term ride as everyone starts piling on. Watch for it. |
| MS Dev Studio 6 and Dev Studio .NET COEXISTENCE | Sun 27 Apr | Codex |
| Can dev studio 6 and dev studio .NET can coexist on the same machine (OS) Or if you try to install dev studio .NET it will overwritte dev studio 6 by updating it ? |
| Sun 27 Apr | JWA | Assuming that you mean Visual Studio 6 and Visual Studio .NET: yes, they can both be installed and live quite happily side-by-side. You can slo have VS.NET and VS.NET 2003 side-by-side, but I'm not sure why you'd want to. Regards, --Josh |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | You can have all three side by side. You'd want the two VS.NETs side by side if you have projects that are targetting 1.0 (VS.NET 2003 cannot explicitly target 1.0, as it always compiles against 1.1, although you can 'ask' for 1.0 by using some configuration options). |
| Code reading | Sun 27 Apr | Codex |
| I would like to read efficient, robust, clean code. Can anyone recommend me some widely available source code to look at ? Maybe a compiler or an OS ? Something which is full of wonderfull tricks and where I can learn some stuffs from |
| Sun 27 Apr | Snyder | Recommending... http://msdn.microsoft.com/net/sscli/ Or... http://www-124.ibm.com/developerworks/oss/jikesrvm/download/index.shtml Or for something smaller, http://www.sweetcode.org/ PSTZZZQ |
| Donald Knuth - The Art Of Programming Series | Sun 27 Apr | Codex |
| Ive read somewhere that the Knuth books have got to be part of the bookshelf of any serious software engineer. So I purchase volume 1 (Fundamental Algorithms) And I find the book not very accessible, its seems that you need to have a really strong mathematical background to learn anything from that stuffs. Can anyone give me some technics, tools, tips to apprehend such material. I would love to be able to understand the book at some point Waiting for your advices :-) Codex |
| Sun 27 Apr | Mike | 'it's seems that you need to have a really strong mathematical background to learn anything from that stuffs.' it's seems that you need to have a really strong mathematical background to excel at a high level in a computer related field. That's just the way it is. Some can do it, some can't. There is no shame. |
| Sun 27 Apr | choppy | Hm. Anyone can do math, it just requires practice. The Knuth books are probably not the best place to start. The Knuth books are really hard. Basically, anyone recommending those books to you is lying; they did not read them. They are on many a wannabe's bookshelf though. (here's a tip for ersatz math/CS wannabes: you'll get further in life with art and photography books on your bookshelf; women won't think you are such a bore). You can get a PhD in computer science from a top school without ever opening one of those books...the example code is in ASSEMBLER, for god's sake. In any case, if you want to get hot and heavy with algorithms, I recommend typing in the code from the sedgewick 'algorithms in C' books to get a practical understanding of common algorithms, and getting the big fat Cormen/Leiserson/Rivest book as a reference. If you get through those two books you'll probably know more than most people with a 'BS' in CS. |
| Sun 27 Apr | Julian | I've read the entire Cormen/Leiserson/Rivest book, and major chucks of The Art of Programming, but it hasn't been very helpful in my work. A thorough understanding of the Java Collection classes, or the equivalent in whatever language you're using, provides a lot more practical value. Few programmers have any reason to write their own sorting, searching, or arithmetic routines from scratch. |
| Sun 27 Apr | one programmer's opinion | Codex, As someone else already mentioned you would be better off purchasing an algorithm book that uses a particular high-level programming language. Theory vs. Practice You seem to have mistaken computer science with software engineering. Donald Knuth is a well-known computer scientist. Most computer scientists are researchers who study theorectical computer software/hardware related topics rather than practical topics. For example, computer science came up with the notion of formal specifications, a very mathematical way of representing software requirements. This mathematical approach is rarely used in practice because software requirements have this nasty habit of changing a lot during the lifespan of a software project. Even though most software developers find formal specifications to be useless this subject is still being taught and advocated within academia. |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | The importance in understanding the core data-types and algorithms today isn't so much so that you can write them, but so that you can select the right one to use (i.e., when to choose an array vs. a list vs. a set vs. a queue vs. a stack vs. ...). |
| Sun 27 Apr | Codex | One Programmer's Opinion : I did read in a sofware enginering book that the formal method has been extensively used for some projects at NASA. Dont know if they're still using it though ... |
| Sun 27 Apr | Michael Kohne | I've got all 3, and I have to say that I haven't done an in-depth read of them, but I have 'skimmed' all three to some degree and I can say they are helpful. I think of them more as a reference than a read (though I would like to read them when I have more time). I've skimmed the sorting/searching book completely, the others I delve into when I need them (I've gotten into the randomness testing stuff a couple of times, as well as some other stuff). I also find them an occasionally useful as 'give me an idea' books - I flip through the TOC and index when I'm stuck and things will leap out and 'unstick' my brain. As to them being Computer Science texts, not Software Engineering books: Well, yes. And our field (Software Engineering) is derived from academic research conduced by Computer Science types. While I wouldn't want to be an academic, I realize that if I don't have a good understanding of the underlying concepts, I'm NEVER going to do my best work. So I read O'Reilly books as well as computer science texts. To put it more simply: Everything we work with is an abstraction. As Joel put it, abstractions leak (some more than others). If you don't understand the underlying stuff, or how the abstraction is built, you can't understand the failures. |
| Programmer's Secretary | Sun 27 Apr | easy |
| Hi, How would you define the task of secreatary of a programer. If manager have secreatary to help them and keep them on the track with information and work and meetings and appointment, how would secretarial jobs would help programmers to finish their task well? easy.. |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | Like a leprechaun: 'fantasy'. :-p Seriously, I've never felt like I needed a secretary as a programmer. |
| Sun 27 Apr | murph | Once upon a time (18 years ago when I worked in a biggish company) we had programmer's assistants who were, more or less, PAs for a team of programmers. In that instance there was enough administrative running around (literal running around to fetch and deliver stuff) to keep them busy. Nowadays... well depending on the team there's probably still enough stuff for maybe half a dozen developers to keep an assistant busy one way and another - keeping minutes for a start, any number of other things. Whether there's a return on the investment is an entirely different question... |
| Sun 27 Apr | Daniel Shchyokin | I would describe it as 'project manager' |
| Sun 27 Apr | Philo | I would suggest that a project team should have a tech writer, and he/she should be amenable to performing some secretarial duties, but otherwise print your own damn memos. [grin] Honestly - I'd be happier with a laser printer on my desk. Philo |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | Cheap laser and LED printers are available now. I suppose one advantage for the company to having a common laser printer is that the laziness factor for engineers means they'll print out less (because they have to get up and get it). :-p I didn't really give it much thought, taking it for advantage, but you're right, I need to have a printer handy. |
| Sun 27 Apr | Julian | I'm used to there being an office manager, who deals with supplies, deliveries, answering phone calls on the general line, and other secretarial responsibilities. |
| Sun 27 Apr | easy | Hi, I infact ask from programmatic perspective. ie.. If are a programmer and you are given the secretary, what type of work you would assign to secretary who can work for you. Programmers often have deadlines and if they have someone manage themselves yet not the boss or manager and who knows common stuff, that would be helpful. Assume you are programmer, what is the best way you can keep track of info, delieverd by someone and relavant to you which can make you a better programmer and speed to accomplish your goals. The one whose reach and vision are beyond computer tools to help you. |
| Installing Visual Studio Installer? | Sun 27 Apr | Herman |
| How do you install the Visual Studio Installer? The msdn documentation doesnt tell you how and it is not installed on my VS 6.0 Enterprise system and I have installed everything. I still cant create a Visual Studio Installer project because I am not given the option. I can create a zip or cab file, but not an Installer project. Maybe I should use a different installation package. |
| Sun 27 Apr | JWA | Hi Herman, Do you mean the Package & Deployment Wizard? The shortcut to that is installed in the 'Microsoft Visual Studio 6.0 Tools' subdirectory of the 'Microsoft Visual Studio 6.0' folder in the programs menu. As far as I know that's the only way to handle an installation project in VS6. It's a bit tedious the first time, but it does save the settings and in the end it works ok. I've had 3 versions of a shrink-wrap VB app selling in for two years using that installer, with few reported problems. There are a few issues with uninstallations followed by renstallations on Win 98 and Me machines, but they are documented in MSDN. Regards, --Josh |
| Sun 27 Apr | Herman | Thanks for responding Josh. I am referring to the actual Visual Studio Installer which you are suppose to be able to run from Visual Interdev or Visual J++. |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | I don't believe that it ships with Visual Studio. I seem to recall having to download it, but I could be wrong (maybe it was the MSI version you had to download). Regardless, using a higher level installation tool is highly recommended. There are some free ones (like Nullsoft) and some commercial ones (like Wise). There's also some diseased ones, like Installshield, that are best to leave be. :-p |
| CVS Server setup in Redhat Linux 9 | Sun 27 Apr | Alai |
| Im trying to set up a CVS server on a fresh install of Redhat Linux 9, but from the FAQs Ive read, I need to execute a cvspserver command ... that I dont have. Ive looked through the CVS website and the RedHat website as well as the package management in RedHat to try to find out how to set it up. I realize that many things are not a matter of Click and its set up, but it seems like its a bit much to have this much trouble setting up whats supposed to be a very common source control system. Im thinking of just installing CVS for NT since that is easy to set up. I figured that I would at least find a FAQ that gave me a step by step for a current distribution. --End Rant-- |
| Sun 27 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | Are you sure you're not missing a space there? 'pserver' is a valid command to give to cvs. |
| Sun 27 Apr | Roland Kaufmann | I don't know if this information still applies to the latest version of RedHat, but the last time I did this, it went something like this:
The 'classical' way to do this is to add an entry in /etc/services that tells the portnumber on which the server is going to run, and I believe that's where the 'cvspserver' comes from:
cvspserver 2401/tcp
and then instruct the superserver to start up the CVS program when someone contact this port, by adding the following line in /etc/inetd.conf (all on one line)
cvspserver stream tcp nowait < |
| Sun 27 Apr | Roland Kaufmann | One more thing; you will have to 'hup' the superserver before the changes in the files take effect, using pkill -HUP xinetd I assume that you have already initialized the CVS repository in the 'home' directory? |
| Sun 27 Apr | Giovanni Corriga | Place this in a file named /etc/xinetd.d/cvspserver service cvspserver { # disable = yes socket_type = stream protocol = tcp wait = no user = root passenv = PATH server = /usr/bin/cvs server_args = -f --allow-root=/usr/local/cvsroot pserver } Change /usr/local/cvsroot to the base directory of your cvs repository. Then as root type 'service xinetd restart', and your CVS server will be ready. |
| Sun 27 Apr | Giovanni Corriga | Warning: my solution is a nightmare from a security standpoint, but it's good enought to give you a working CVS server. Take a look at the Cederqvist http://www.cvshome.org/docs/manual/cvs.html to learn how to secure your CVS server. |
| Smoking | Sun 27 Apr | suhu |
| I used to be on off smoker. By on meaning a couple of sticks a week and stuff but lately I seem to be smoking a lot (one pack a day). Have not been getting to much work lately (freelance programmer) and have too much time on my hands for my own good. Any of you been through this and managed to cut down or give it up for good. thanks in advance for any suggestions. |
| Sun 27 Apr | choppy | hey. smoking is rough. i used to smoke 2 - 3 packs a day. how i gave it up... is i moved to florida for about 4 months. i've read that the best way to quit smoking is to change your environment. it is easier to quit cold turkey if you leave your normal routine for a while. i lived in NYC...and i smoked when i walked around. then i moved to tokyo...where EVERYONE smokes...AND smokes were like $2.00 a pack.. then i decided i was probably going to die at age 35 if i kept up the smoking... so i found a contract in florida... and just went to the gym in the AM and the beach after work... and managed to cut it out. now in a bar i literally almost barf if offered a smoke... so... i dont know if it is practical for you to change your environs... but if you can, i would do so. truly, smoking is enjoyable...but horrible for your body... so if you can... move to the beach for a while...give it up. you'll never go back |
| Sun 27 Apr | Stephen Jones | I smoked four packs a day for thirty-two years and gave up fifteen months ago. The change of environment in this case was an intensive care unit on a drip feed with anything by mouth banned for a week. I couldn't be bothered to take up smoking again afterwards. But I have gained nearly 20lbs weight, and actually feel no healthier (in fact slightly worse) than when I was on four packs a day. Giving up does has an effect on carpets, furniture and clothes though, as they no longer are covered in cigarette burns. |
| Sun 27 Apr | Just me (Sir to you) | Analyse where and when you smoke. Avoid smoking at places or during activities where you spend most of the day (e.g. at the desk, on the couch etc.) Don't get into the black and white thinking (You are either a smoker or you must quit 100%) This polarization is extreme and unproductive. Enjoy a few smokes in 'special' circumstances, but do not get into just smoking routinely. |
| Sun 27 Apr | S.C. | Quitting smoking is easy. I've done that many times... |
| Sun 27 Apr | The Real PC | [and actually feel no healthier (in fact slightly worse)] This is wrong wrong wrong. Either you don't exercise at all (and that is practically guaranteed to make you feel bad) or your lungs were so damaged from smoking they did not recover. But as far as I know that isn't possible, and they always can recover somewhat. Everyone MUST exercise -- we are animals, not plants, and circulation will not function correctly if your muscles never move. In addition to exercise, you have to re-learn how to breath. Smokers forget how to breath and their lungs have barely any capacity. You smoked for such a long time you may have completely forgotten how to breath and are surviving on a bare minimum of oxygen. This is no good for your brain and will prevent you from thinking clearly, and your programs will be full of bugs! |
| Sun 27 Apr | Ron Porter | I agree with the 'change environment' comment, but would add that 'change routine' or 'substitute healthy habit' may be just as (or more!) effective. The way I did was to do some exercise every time I felt like a smoke: Have breakfast: 10 min. stretching exercises Coffe break: situps Lunch: walk around block Coffee break: chin ups Supper: bike ride Sit around watching TV: more stretches Before long I was almost as addicted to these healthy activities as I had previously been to cigarettes. Bonus result: I'm the only person I know that *lost* weight as a result of quitting! Good luck! |
| Sun 27 Apr | Stephen Jones | Dear PC, Why not exercise your brain for once and give your fingers a rest. You may have decided that anti-smoking is the new religion, but you know nothing about my lungs, my health, the amount of exercise I take or anything else. The day I want a remote diagonosis I'll consult an online astrologer. He can't know any less than you. |
| Sun 27 Apr | Phil | I smoke a few sticks when I want to (usually after some food) and also have quite some free time available when enough cash is on my account. I choose to go to the gym when I've nothing special to do (or read books, or call prospects). The results are quite good: - I met new friends - My body is in much better shape - I can concentrate on programming tasks better - My willpower has been enhanced significantly But I keep on smoking some sticks. But I do not want to be under the control of nicotine or any other thing (especially a boss). You said you are a freelancer, so I guess that freedom is also an important value for you. |
| Sun 27 Apr | The Real PC | Stephen, It does not require any psychic powers to know that a person who smokes 4 packs of cigarets a day for 30 years has damaged lungs! If you did not feel any improvement after quitting, something definitely is wrong. But I have another theory -- you actually do feel better but you hate us anti-smoking fanatics so much you won't admit it. |
| Sun 27 Apr | choppy | nicotine is a stimulant and antidepressant, so when you quit smoking the chances are likely that you will feel very sluggish. you certainly do not immediately feel 'better.' that is one reason i chose florida to quit cold turkey. i tried to quit many times in the northeast and could not because i needed some fix to stave away the winter blues. smoking is a weird beast, certainly there are obvious health issues. But, i know a lot of bike messengers, and they all smoke like chimmneys. |
| Sun 27 Apr | Eric Debois | Im smoking now but a few years ago I managed to stay quit for a year. (I started again because of a change in environment and life circumstances.) Well, the way I quit that time was quite weird. Being young and reckless I ate some psychedelic Magic Mushrooms when the opportunity presented itself. That in it self was a strange adventure (ordeal might be a better word), and during the intoxication I tried to smoke. The mushrooms had made med tremendously sensitive though, and the experince of pulling poisonus smoke into one of the most delicate organs in the body was excruciating. I felt so sick from it that it took 6 months before I even considred having a smoke again. I decided att that point not to pick it up again.. The memory of the experince faded however. When I found my self in a situation where smoking offred some comfort as well as quick way of finding freinds in a place where I had none, I started again. I have actually considred getting hold of some mushrooms again just to help me quit, but I wouldnt know where to start looking. For better or worse. |
| Sun 27 Apr | Bored Bystander | >> Being young and reckless I ate some psychedelic Magic >>Mushrooms when the opportunity presented itself. >> during the intoxication I tried to smoke. >> The mushrooms had made med tremendously sensitive though, and the experince of pulling poisonus smoke into one of the most delicate organs in the body was excruciating. I felt so sick from it that it took 6 months before I even considred having a smoke again. This sounds like the Ludovico treatment in 'A Clockwork Orange'. (IE: the scenes where Alex is strapped down in a theatre watching sex and viloence movies and being given nausea drugs in order to make him associate barfing with 'uh, the old in-out, luv.') But a self administered Ludovico treatment is a new one on me. Hey, whatever works. :-) |
| Array to worksheet... fast | Sat 26 Apr | #If |
| Anyone know of a faster way to fill an Excel worksheet with an array other than iterating the Rows/Columns? Thanks |
| Sat 26 Apr | njkayaker | Paste it as tab-delimited text into the sheet (you can do this programmatically). |
| Sat 26 Apr | Duncan Smart | You can assign the array to the range! |
| Sat 26 Apr | Duncan Smart | ... that's if it's a VB-style (variant) array. |
| Sun 27 Apr | Adrian Gilby | Don't forget to turn off window updates while the copy is in progress. It's been a while since I did Excel VBA, but I recall it made a huge difference. |
| Sun 27 Apr | #If | Thanks guys... works like a chucking farm! |
| WiFi in EMF sensitive environments | Sat 26 Apr | David Roper |
| Can anyone tell me whether WiFi networking is permissible in hospital wards? My local acute hospital insists that everyone turns off their mobile phone one entering, the published explanation being that the signals that phones regularly send to inform the network as to which cell they are in can interfer with monitoring and other equipment. This is the same explanation as given by airlines and may just be precautionary rather than a serious threat to safety. Even so, if its true for phones, is/will it also hold for WiFi frequencies/energies? |
| Sat 26 Apr | www.marktaw.com | If it's in the same frequency band, I'd steer clear. If it's in different frequency band, I'd steer clear. |
| Sat 26 Apr | Stephen Jones | We have some old, old language labs and we play tapes from a tape deck throuigh two wall speakers. If the students have the mobile phones switched on, even though they are not being used, the interference comes through the speakers and makes the tape uniintelligible. When I was at colllege all the police walky-talkies used to come through the stereo. Great if you wanted to avoid being busted! There are some reported cases of input put on HP wireless keyboards being inputted on computers five or ten houses down the street. I have two Satellite TV decoders that come from the same manufacturer, though they have been customized for each separate digital package. There is no way I can remotely change channel on one decoder without digitally changing it on another. |
| Sat 26 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | 802.11a almost certainly is not interfering with anything, as it's 5GHz. 802.11b and g, on the other hand, live in the rather noisy and crowded 2.4GHz band. Cell phones do not operate in either band, though. |
| Sat 26 Apr | choppy | david, the hospital where I work does have a WiFi network in the wards. |
| Sat 26 Apr | raindog | Re: cell phones in airplanes: There's no evidence that cell phones affect plane equipment. The main reason they ban cell phones is purely economic: the airlines receive a cut of the revenues from the telephones installed onboard. http://zdnet.com.com/2100-11-501431.html?legacy=zdnn |
| Sat 26 Apr | Michael Kohne | You need to work closely with the hospital's biomed department. I've worked on medical equipment before (an EEG monitor called the Neurotrac II). Depending on the equipment, some of it (especially EEG gear) can be VERY sensitive to noise. And radio signals can do all sorts of wierd and unexpected things inside a building, some of which can cause a very high-frequency signal to alias down into very low bands and interfere with things that you wouldn't expect. In short: Have a long heart-to-heart with the hospital's biomedical engineering folks (the ones who put a 'safe for patient use' sticker on every piece of equipment). They can give you the best advice. If it's feasable, consider avoiding wireless unless you are sure what sort of gear is going to be in use. My one, true rule for dealing with medical gear: Would you let them put you in that ward/use that thing on you? If not, you've screwed up - start over. (Yes, I did have my EEG done by our in-house people. I also had an evoked potential study done by our folks, using our gear, as part of evaluating the equipment.) |
| Sun 27 Apr | Robby | I think the cell phone rule on airplanes is an FCC rule, not an FAA rule. It's not designed so much for interference with flight systems, as it is to prevent problems or 'free calls' on cellular networks when you're up at 30,000 ft. From there, you're within line-of-site of thousands of cellular towers. I've read reports of people not getting charged for cellular calls at that altitude because the cell-switching software couldn't keep up with the speed with which calls were passed from cell to cell. |
| Sun 27 Apr | Robby | On the hospital issue (which was the original question anyway), I would hope that all electronic hospital equipment is tested to be resistant to any publicly available frequency spectrum. Since WIFI operates in such a frequency theoretically those devices should be OK. |
| Sun 27 Apr | David Roper | Thanks everyone. I would like to think that critical hospital equipment would be built to be as immune to RF interference as possible, not least because all those CRT screens, fluorescent lights etc. etc. are themselves RF emitters. You'd hope that any analogue cable runs used shielded cables and that digital protocols would be robust against the occasional interference induced glitch. What caught me by surprise recently was a fault with my ISDN connection. When British Telecom came to fix it, it transpired that their standard cable doesn't use shielding around each pair to avoid crosstalk. Instead, each time they step from one run of cable to another they just switch around the conductor pairs so that crosstalk induced between between different cable pairs tends to cancel. Well, this may be OK for a voice circuit as it'll just get a bit more hiss and crackle, but it sure doesn't work for digital unless everything else is perfect! |
| rigerous tech interviews | Fri 25 Apr | Scott |
| I recently went through a rigerous technical interview, around 3 and half hours( I got the position), where I was asked pretty much everything cs related from OOP,data structures, algorithms, cryptography,OS,threads,sockets, shell scripts, automata theory, reg expressions, database queries, C++,Java, pretty much everything. The questions both tested knowledge as well as doing problem solving given a problem. I found the whole thing pretty draining, being asked around 50 questions in all, can other people share some of their experiences, have people been through these types of all inclusive inteviews for soft. dev positions? |
| Fri 25 Apr | anon | You don't want any job where you don't get interviewed like that. Think about it - would you rather work with people who know their stuff, or who know what their biggest weakness is, and where they see themselves in five years? |
| Fri 25 Apr | Curious | What was this position for? I have never had any interviews like this or even close to this. I also have never used most of what I know. Do you live in the USA? |
| Fri 25 Apr | SteveM | Well congrats on getting the job, personally I would've told them where to stick it! (Assuming I could afford to ;-) If they make you go through that before they'll offer the job, what the hell are they going to be like to work for? My guess is we'll find out soon :-))) |
| Fri 25 Apr | Scott | The position is in the states, small company, around 7-8 software developers. The owner of the company has some software patents, and most of the developers are 'toppers' from Berkley, stanford etc.... I seem to be the most underqualified among all of them. But I don't know why I was asked such a variety of questions. |
| Fri 25 Apr | Scott | Oh.. I also had to give 5 -6 samples of C++ code |
| Fri 25 Apr | Curious | Well to get a patent takes almost nothing except filling out paperwork. You don't even need a working product just the idea. I worked with a guy who had 80 patents...of ideas only. I am still curious to know what area you will be working in, is there a product involved? |
| Fri 25 Apr | Scott | A few products, mostly regular applications, ERP systems, some algorithm development withrt. to info retreival, mostly business applications in C++,Java |
| Fri 25 Apr | Bored Bystander | >> A few products, mostly regular applications, ERP systems, some algorithm development withrt. to info retreival, mostly business applications in C++,Java And they grilled you like they were looking for Stephen Hawking's clone. I know that a lot of techies will say 'great! a really meaningful interview where the interviewers really know their shit, and they hand picked the absolute best candidate!' The 'appeal to geek vanity' approach. I don't agree with this thinking at all. Again, I say that tests and interviews like this are inherently flawed. They are associated with a presumptions that verbal mastery of the subjects is equivalent to ability to perform. Mainly they're a self indulgence of egotistical companies with too much time on their hands to waste expressing their presumed superiority with a bunch of BS questions. It *could* be that such an interview process rejects several candidates that may be superior picks but who don't verbalize answers well or think on their feet in a test situation. The job's going to be writing code, not taking tests. Why is it that so few techies see this, and the most gifted techies are seduced by a surface impression of a 'meritocracy' that is basically just someone's huge ego at work? It's biz-apps, not primary research into cryptography, for pete's sake. And not having the street smarts to see this is a failing of most geekly types. (Scott, I'm not saying that the job isn't worthwhile or the employer's intentions are not honorable, I'm just saying that the stuff that people raise up as a 'must have' often never even gets used in real life.) I guess it's similar to the way that the prettiest girls in high school treat guys like crap and make them jump through ridiculous hoops. Act like you're unapproachable, set up some process moronically obtrusive for a 'standard', and become highly desirable.... |
| Fri 25 Apr | Scott | That's what they kept telling me, they are extremely selective... although I don't know if this kind of tech. interview is the way to weed out applicants. I was asked to describe Dijstra's shorteest path algorithm, and Kruskal's algorithm from graph theory, I did not even remember them , and I'm sure most techies won't. |
| Fri 25 Apr | Curious | I have to agree with Bored Bystander. I hope you actually will be using a tenth of what they quizzed you on and that your coworkers truly are 'toppers'. Anyway, its a bad market so be happy you at least have a job. It will be a learning experience in IT and real world business. |
| Fri 25 Apr | Dimitri. | Ha ha ha sounds like fun. SteveM said it exactly like I would. I would guess that the owner (or the interviewer) is some academic that just wanted to see if you know what 'a good programmer SHOULD know'. No disrespect for academics, being one myself, mostly, but some can get rather religious about things of no real importance (like computer 'science', for example)... We where once open for a PhD position. Lots of applications, all possible people. We finally sorted them down to five people. One of them had finished with top grades (and I mean, top, they where all max) but when we asked him if one particular area sounds interesting, he said, 'shure, that is interesting'. And what about that other one? 'Shure, that too'. And so on. And another guy had medium grades, some where top but most where medium, but he had really interesting ideas and was passionate about certain problems. And when he got started he could really go on about something. And yes he was smart. Can you guess who we hired? You're wrong. We got the girl with the big tits. No, just kidding. There was no girl with big tits among the applicants, so we settled for the guy with medium grades. So the conclusion is that I'm wasting my time like an idiot in cs when I should be out going to movie director school right now. |
| Fri 25 Apr | Philo | A senior manager at one place I worked was proud of one interview question - 'Name the layers in the OSI model' He said 'only one person I ever interviewed got it right' My reply was 'maybe that's more an indicator of the value of the question than the applicants...' Philo |
| Fri 25 Apr | | Interviews like that are about boosting the ego's of the hirers, who will generally be recently out of, or closely tied with, formal learning experience (uni) where they've done all those questions in exams. Your comment about you being the least qualified sums it up. That's the way they want things to be; that you know they're the best. That approach is actually 1-dimensional and, in my considerable experience, those outfits don't generate good product. They get a first round of funding OK. But not a second. |
| Fri 25 Apr | Prakash S | Bored, Interesting reply, how would you interview a person for..lets say a software development position. |
| Fri 25 Apr | one programmer's opinion | I have been through all types of interviews. * The group interview * The meet the department manager interview * The meet several people in a row interview * The meet the HR/Recruiter interview and then take a battery of online tests interview. * The meet the HR rep but don't take any tests interview. Note: had one guy gall to tell me 'I can't hire you but I just had to meet you' * The interview with the team lead and project manager interview However, I have never had an interview where I was verbally grilled for over 3.5 hours on technical issues! That said, I don't have a problem with this type of interview as long as most of the questions that were asked during the interview truly apply to what you the job candidate will be doing once you get hired. |
| Fri 25 Apr | Scott | Doesn't microsoft do something like this in their interviews, or so I heard? |
| Sat 26 Apr | Bored Bystander | >> Interesting reply, how would you interview a person for..lets say a software development position. I'd ask the person to submit to one of the following tests when they interview: - Implement the solution to a coding problem on the spot. They would be given a goal, or they would be asked to formulate a goal, and would be asked to develop a solution in a given time frame. They would have access to a PC loaded with the tool du jour. - or - - Bring in compilable source code on some media. Ask them to load it into the development environment, and demonstrate the code. Ask them to discuss any design tradeoffs or unique problems encountered during development. Ask them to walk through debugging of any interesting functionality in the code. The point isn't to stress them. The point is to see if they can actually do the work and can talk intelligently about it. The only way to really know is to ask the candidate to demonstrate their craft. A secondary benefit of this would be to determine if they can work with the interviewer. Some programmers (as well as other IT niches) have this highly defensive personalized 'thing' against any sort of actual code demonstration. They claim that it's like being made to 'perform' like an organ grinder monkey. I think that viewpoint is erroneous, because it's quite possible to BS as though you know the material, yet be effectively unable to synthesize a result from the raw materials. Actually, I think that companies hiring in this field with a 'black and white' mentality are in error. They think that they have to make a multi-year commitment based upon a few conversations. Why don't companies do the intelligent thing and try the candidate out on a moonlight basis on contract for a couple of weeks? Everyone wants to jump off a cliff, which results in total anality of the decision making. Instead, take the pressure off and do a trial run of the new hire, with no make or break decision made based only upon talk. |
| Sat 26 Apr | Prakash S | Makes sense, thanks. |
| Sat 26 Apr | Simon Lucy | 'Name the layers in the OSI model' Bum I could answer that question. Those kind of interviews strike me as more a lack of being able to conduct a conversation than anything else, a cast of mind that there are verifiable objective facts about a person which make them more likely to fit in. In reality, if you get to the interview people's minds are made up within around 30 seconds. |
| Sat 26 Apr | M.M. | Last interview, i had to write a bit in C (with pointer + bit arithmetics). I had to write stuff on paper - and i was told that the stuff must compile correctly and that it has to work. Then they took my sheet of paper - and returned it (you got errors!). Don't know what they did with the stuff - i didn't see the reviewer, just the girl from HR Needless to say that i felt like a piece of shit afterwards. Anybody with a similar experience? |
| Sat 26 Apr | Philo | When I'm being interviewed I generally make anecdotal comments about the topic I'm being interviewed about. The kinds of jokes/observations/gripes that you generally only make when you've worked in the environment. For example, once an interviewer asked me 'what is the default timeout on IIS?' I answered '900 seconds, but I always shorten it to 45' 'How come?' 'I used to forget movenext's a lot' The interview was effectively a pass at that point, and I got the job offer. As an interviewer, I try to ask the same types of questions - as Joel points out, I want to know as much about personality as knowledge, and I think you learn more about experience by looking for anecdotal comments as from grilling. Philo |
| Sat 26 Apr | Curious | You guys must be doing some interesting programming. I have never been asked or even come across any of the above comments in the real world. I just contract for large companies mainly doing ERP and datawarehousing; it's incredibly dull but you make so much that it's not worth doing the hard stuff. |
| Sat 26 Apr | one programmer's opinion | Bored Bystander wrote, 'Implement the solution to a coding problem on the spot...' My guess is that some employers already do what you are advocating. However, I bet those that do only offer one type of coding problem test. While it seems to me that it wouldn't be too difficult to create a 'you have 2 hours to build me a GUI based tool program' type of test such as -- create a calculator program -- I am not sure how fair such a test would be if the person being asked to take the test claims to have mostly server-side development experience or specializes in some other aspect of software programming. Bored Bystander wrote, 'Bring in compilable source code on some media. Ask them to load it into the development environment, and demonstrate the code. Ask them to discuss any design tradeoffs or unique problems encountered during development. Ask them to walk through debugging of any interesting functionality in the code' I like this idea and I also like the idea of allowing for 'flexibility' or more than one way to prove that you know what you claim to know. Also, I agree with the contract-to-hire philosophy as well. |
| Sat 26 Apr | one programmer's opinion | Hi Curious, If you don't mind could you explain how the interview process typically works for you. Money wise, ERP and data warehousing work is great stuff to be doing. However, not too many people have the knowledge and appropriate prior work experience to be offered this type of work. I am assuming that you send your resume to various staffing firms and they simply checkout your past clients/employers and then submit your resume to the corporate client looking for help. |
| Sat 26 Apr | Curious | If you don't mind could you explain how the interview process typically works for you. There isn't one. Money wise, ERP and data warehousing work is great stuff to be doing. However, not too many people have the knowledge and appropriate prior work experience to be offered this type of work. If you know half of what was discussed on this thread then you know more thant 90% of ERP / datawarehousing 'professionals'. |
| Sun 27 Apr | B | That is how I have been interviewed and how I interview now. I dont see a problem with this method. I've probably been interviewed 15 times and have interviewed about 40, so far so good. Bonehead hired 0. Heck, I didnt take a company seriously if they didnt offer a hard interview. But then again I am a hardcore programmer, with the ego to match. |
| Sun 27 Apr | | 'what is the default timeout on IIS?' That sounds like a question tailor made for MCXX 'pros'. |
| Sun 27 Apr | T. Norman | 'Bonehead hired 0.' Ok, you hired no boneheads, but of that 40 how many people did you hire? I think strong technical interviews are a good idea, but if they are so ridiculous that nobody can survive them -- especially if many of your own employees couldn't survive them -- they can cause you to miss out on hiring good people. |
| Sun 27 Apr | Scott | how much technical is too technical? |
| Sun 27 Apr | Philo | 'what is the default timeout on IIS?' You should know it if you do web development, since it's the number you have to change because it's too damn long. It's an indicator as to whether the interviewee has ever dealt with IIS on their own machine. And from my perspective, 'I don't know, but I know it's too long so I always change it' might be an acceptable answer. BTW, I never asked that question, and I don't think I ever would, but it's the kind of question that shows if someone's done more than simply open visual studio, type, compile, run. Philo |
| Sun 27 Apr | | Maybe it's too damn long because you don't know how to properly program high performance web applications? |
| Sun 27 Apr | | Ok, maybe that was a little harsh, but my point is that the timeout is not a one-size-fits-all setting. It's naive and senseless to say that it's 'too damn long'. Many moons ago I worked on a project that never changed the default timeout. Why? Because it was an internal client/server app where both parts were custom coded. It was never an issue. An 'inactive' client app still sent an ack, and we always terminated our connection when the app exited. |
| Sun 27 Apr | realist | My time is far too valuable to set aside 3-5 hours for an unknown ERP application developer with some unprofitable patents. I would designate 45min - 1hour to talk to this type of organisation. |
| Sun 27 Apr | Philo | Heh. [candidate checks watch] 'Are you guys done yet? I'm getting concerned about an organization that can't judge an individual's talent in under two hours.' Man that would take some cojones... Philo |
| Bigint lib (up to 300^60000) | Fri 25 Apr | na |
| I need a library or some math intro how to handle integers up to 300^60000. Any idea? |
| Fri 25 Apr | Roel Schroeven | One possibility is the GNU MP library (http://www.swox.com/gmp/). |
| Fri 25 Apr | Li-fan Chen | Problem with most large number (large integer big integer) libraries is that they don't have a ActiveX wrapper around them. .Net SDK and Java both have large number libraries. -- David |
| Fri 25 Apr | S.C. | No matter how smart your bigint library is, you just cannot handle numbers that large, considering the total number of atoms in the the universe is about 10^80. If you are talking about arbitrary numbers, this is unrealistic neither in terms of space complexity nor time complexity. If you are dealing with special numbers only, like those with most digits zeros, of course it can be done. But I think you have to write your own library. |
| Fri 25 Apr | na | cool, seems to work fine. thx |
| Fri 25 Apr | na | s.c. even if there are 10^80 atoms in the universe it doesn't mean that you cannot represent it with a few chars. in this case 5 chars: 10^80 |
| Fri 25 Apr | | If you want ot represent it in chars, why do you want a big int library? I know, I know, in C char and int can be equivalent. Maybe instead of a big int library you could just write the chars in a large font? (Okay, its Friday, and this is getting silly) |
| Fri 25 Apr | Frederik Slijkerman | You cannot represent 10^80 different values without having at least as many bits to store a value in. Assuming that you need one atom per bit, etc... |
| Fri 25 Apr | Just me (Sir to you) | Common Lisp has this built in. You could use Allegro ( http://www.franz.com/products/ ) but there are easier solutions. http://www.codeproject.com/csharp/BigInteger.asp (never used this myself though) |
| Fri 25 Apr | Mike McNertney | na: Representing the number '10^80' and representing 10^80 different values are very different things. As you said you can represent the number in 5 characters, but to represent that many different values you need far more information. However, Frederik was wrong on an important point. You don't need 10^80 bits to represent 10^80 values. You need log2(10^80) bits. This is only around 266 bits. To represent 300^60000 different values, you need log2(300^60000) bits, which is only about 500,000 bits. |
| Fri 25 Apr | Eric Lippert | Frederik, I think your math is not quite right. The number of bits required to store any integer from 0 to n is the base-2 log of n. 10 ^ 80 < 16 ^ 80 = 2 ^ 320. So 320 bits would more than do it. 300^60000 < 512 ^ 60000 = 2 ^ 540000, so you'd need around half a million bits per integer. That means that adding two integers would take a little less than one meg. I hope you've got a lot of ram. Now I'm curious -- what's the application? Eric |
| Fri 25 Apr | Mike McNertney | Furthermore, even without the conversion to binary, it is pretty silly to claim that you can't represent 10^80 different values, considering you can do so with an 80 character string of digits. ;) |
| Fri 25 Apr | Eric Lippert | Jinx. |
| Fri 25 Apr | Gareth McCaughan | To clear up the 'number of atoms in the universe' thing: To represent 10^80 different values at one bit per atom you need log_2(10^80) atoms. That's less than 300 atoms. If there are 10^80 atoms in the universe and you encode one bit into each atom somehow, then the largest number you can represent will have 10^80 *bits*, so it will be about 2^(10^80) in size. That's a lot bigger than 300^60000. To represent numbers up to 300^60000, you need log_2(300^60000) bits, or about 500k bits, or about 60k bytes. |
| Fri 25 Apr | Gareth McCaughan | Ahem. I'd like to mention that the last 5 replies weren't there when I followed the 'Reply' link. :-) By the way, anyone following the suggestion to use Common Lisp might want to consider using the implementation called CLISP rather than any of the commercial ones if bignum operations make up a lot of the runtime. CLISP's bignum implementation is based on GMP, and it's faster than any of its rivals. (GMP is, I think, the fastest bignum library around.) On the other hand, because CLISP doesn't compile to native code it's generally the *slowest* Common Lisp implementation for everything other than bignums. :-) (Python has fairly well integrated bignums too.) |
| Fri 25 Apr | Dennis Atkins | SC & Frederick, I think you're confusing matters. 10^ 80 = 100000000000000000000000000000000000000000000000000000000000000000000000000000000 So, representing and working with these numbers is no problem at all. What you can't have is a computer with 10^80 bits of memory. But that's something entirely different. Your own computer has probably 8 gigabits of Ram, right? But you are still able to work with numbers larger than 8 billion... |
| Fri 25 Apr | Andrew Murray | Please pardon me if this is ignorant, I'm just a student at University, but doesn't bigint storage in actuallity require far more than log2(n) bits in actuallity? One would think that in order to implement truly unbounded integers in a archetecture independant way, one would need to use a list based data structure (I'm currently writing my own math library using this technique, just to learn). Thus, for each digit, one needs space for the number and a pointer to the next digit... Am I wrong? Is there a better way of doing this? |
| Fri 25 Apr | Steven C. | Well to be sure you need some overhead, but you could also do it array style: one int which tells us the length of the array, and then an array of bytes, with the last byte 0 padded as necessary. Interpret this as a number by appending them (in whichever endianness makes you happy). or however you want to implement it. |
| Fri 25 Apr | Andrew Murray | Steven, But what if, on any given archetecture, the length of the unbounded integer in question is greater than the largest integer size? |
| Sat 26 Apr | S.C. | You guys are right. I guess I was thinking about certain particular problems when I made that stupid mistake. The largest known prime number is (2^13466917)-1 which has 4053946 digits. This was found by Michael Cameron and other guys in 2001. It took an 800 MHz AMD T-Bird 42 days. So I think 300^60000 is quite safe. And good luck with your big ints. |
| Sun 27 Apr | Frederik Slijkerman | Apologies for the stupid mistake... :-) |
| Good Commercials | Thu 24 Apr | Albert D. Kallal |
| The recent discussion about good commercials raises a interesting question: Why do companies not make these good commercials more readily available? Wendy’s commercials, and even some from Hertz rent a car are classics. Yet going to those sites does provide any links for viewing of commercials? I am sure we would all love to view those Apple Commercials. Especially those classes like “1984” super bowel. Of course, the IBM’s commercials are also terrific. They can be viewed at: http://www-3.ibm.com/e-business/doc/content/ondemand/tvspot.html Those IBM commercials were just put up last week, and no doubt this was due to requests. For sure, any good move site has trailers that are of course commercials. Sites like Star Wars and the matrix fully realize the consumer pent up demand for commercials. And, many even try to out do each other. For the example the new matrix trailer is whopping 100 megs in size, and 1024 across in size. This is quality far beyond what you get with dvd. You also need at lest a 1.2 ghz pc or better to even view such a large trailer. ( there of course smaller ones available too). http://whatisthematrix.warnerbros.com/ It seems so obvious to make commercials available? I love watching them, and to feed consumers desires to view good commercials seem very logical to me. I think companies are not using their web sites well when then fail to include good commericals. Albert D. Kallal Edmonton, Alberta Canada Kallal@msn.com |
| Thu 24 Apr | Philo | Can you believe that companies refused to pay adcritic to run their ads? Adcritic was a top ten site when they were around, and people went there voluntarily (and sent their friends there!) to see good ads. If Adcritic was charging a reasonable price, I'm sure it would be cheaper than Superbowl time and reach more eyes. Philo |
| Thu 24 Apr | Spam | Just a thought, but is it possible that the purpose of the commercial is solely to get you to the website, and once you're there, the need is gone? I don't think so, but it's a thought. Advertising is a pretty strange subject though. I recently watched one documentary about the use of sports stars in advertising, in this case Michael Jordan appearing in Nike commercials. The researchers in that situation had deduced (somehow?) that Mr. Jordan himself had little direct effect on sales, but rather his endorsement was a moral booster for Nike staff, and especially Nike salesmen. That sounds suspect to me, but after the success of Google, who can really predict the usefulness of advertising? |
| Thu 24 Apr | Philo@shamelessSelfPromotion.com | My current theory is that competitive advertising (coke vs. pepsi) is a waste of money - people have a preference and are unlikely to be swayed by evangelism. OTOH, *expository* advertising can be worthwhile - showing that a new product is available, or has new features. The funny thing is that I believe more money goes into the first kind, since established brands and products have more money to spend... Philo |
| Thu 24 Apr | S | As far as I can remember from a course I took back in college, there are 2 main reasons why you do commercials: - inform people about new products - remind people about your product. Coke and Pepsi would fall in the second category. There seems to be a need for people to be reminded that their products are cool; otherwise, they just stop buying it. Unfortunately, I can't give you a good example of a product that stopped doing commercials and tanked afterward. Although I have a vague suspicion that Clearly Canadian died because of it and Snapple (who hasn't put out a commercial in a few years) is in the same boat. Also, many commercials are targeted at kids because kids haven't made up their mind yet on which one they like best and can be heavily influenced by celebrities. |
| Thu 24 Apr | Kyralessa | 'Unfortunately, I can't give you a good example of a product that stopped doing commercials and tanked afterward.' That's a very funny comment if you think about it. :) |
| Thu 24 Apr | Jason | Why do tobacco companies bother to advertise? They must spend hundreds of millions a year. I've always suspected that if some brave CEO were to cut advertising, his expenses would drop drastically, and his revenue would only take a minor hit, really bumping overall profits. Not that I want to encourage smoking... |
| Thu 24 Apr | Devil's Advocate | Philo - (Re: Competitive advertising) That just reminded me of the situation about 10 yrs ago ('94, I think) here in the states where the price of almost every kind of breakfast cereal jumped 25-50% in about 3 months. There was a Congressional investigation alleging collusion and price fixing, but it turned out that the big 3 cereal companies were simply laying the advertising on thick and passing the costs on. IIRC, that's about the time I started seeing a lot of generic breakfast cereals in my supermarket. |
| Thu 24 Apr | Philo | S- I still believe that people don't 'forget' about mainstream products. Are people really going to forget that Coca-Cola exists? A counter-example: Tivo has radically restricted their advertising budget, because they've found their best salespeople are their customers and they don't *need* an advertising budget. So watch Tivo and see how they do. Philo |
| Thu 24 Apr | Albert D. Kallal | > Can you believe that companies refused to pay adcritic to run their ads? Adcritic was a top ten site That was one of my favorite sites (as mentioned, I do like commercials). I even have about 30 or more them that I have saved on my computer. A large number of them came from AdCritic. I suspect that companies did not want to have to pay AdCritic. After all, does Pepsi want to pay AdCritic for running a 2 year old commercial? I think the difficulty was that companies could not agree on how, or what kind of terms that AdCritic wanted. However, AdCritic still exists, and is now a pay per view. Interesting, but it seems that many people are even willing to PAY to watch commercials. If AdCrita can survive, then what a great opportunity they have taken advantage of. Albert D. Kallal Edmonton, Alberta Canada Kallal@msn.com |
| Thu 24 Apr | pb | ads.com made a go at it for awhile. I suspect the streaming is pretty resource intensive and the economic case tenuous. |
| Thu 24 Apr | Prakash S | i like the nike/ addidas ads. also my favourite add is the Roberto Baggio/ Johnnie Walker add - can't find it anywhere.. |
| Thu 24 Apr | Prakash S | Jason, I had the same thoughts, but for comercials of Coke & Pepsi. If they stopped advertising, I am sure they could cut the price of a Coke/Pepsi - by quite a bit. Not sure why they don't do it, guess there is some explanation to it. |
| Fri 25 Apr | Simon Lucy | Commercials are available http://www.boardsmag.com/screeningroom/ |
| Fri 25 Apr | Joe AA. | One line of thought says that advertising is only necessary for those things you don't really need. |
| Fri 25 Apr | John Topley | Perhaps they don't make them available because they don't want to generate consumer interest in last year's model? |
| Fri 25 Apr | Slumming IT | I could quote reams of Bill Hicks material, but I won't. You know where to find it... |
| Fri 25 Apr | Slumming IT | That said, there's a really good Honda ad running in the UK at the moment, it's about 2 minutes long and had me completely transfixed - 'isn't it great when things just work'. |
| Fri 25 Apr | John Topley | Those Honda adverts are brilliant! The best adverts that I've seen in a long time. They're even better because almost all of it was done for real, with virtually no CGI. Apparently the campaign cost £750,000 and took six months to set up and film. |
| Fri 25 Apr | Dimitri. | My humble thoughts about advertising... more specifically, why coke and pepsi run so much of it. The thing with that type of adds (coke, cigarettes, etc) is that there is a lot of _image_ involved. Same about a certain class of cars. Now, if coke would stop advertising, of course you wouln'd foget about it, but what would be the difference between a can of coke and a can of 'American Cola' or 'CubaCola' or whatever funny name you may find out there. Again, if BMW stopped running its advertising, after a while it wouldn't be 'wow, cool, a bmw' but more of 'a, hey, isn't that a bmw, I heard they make good cars!' Advertising serves many more purposes then just 'make a product known'. Brand image and so on. You go to a bar. They have a new vodka called 'Keruga'. You have never heard of it. What kind of cheap venezuelan liquor could that be? I'm not gonna buy that. I'm gonna buy Absolut 'cause it's cool or Smirnoff 'cause it's the real thing. Or maybe Stolichnaya 'cause it's the REAL real thing, but they don't run any adds so people don't know so they don't buy it. And so on. Thanks for making me aware of adcritic. |
| Fri 25 Apr | mark | The greatest ads I've ever seen are by Nike, print and video. They must have a marketing-genious tree or something over there. Not as good as they used to be though, maybe the tree burned down. But the HONDA ad that you guys are talking about can be found here. and it's pretty good too: http://www.winisp.net/bogusboy/hondaad.htm |
| Fri 25 Apr | Gregg Tavares | In Japan most companies put their commercials on their website http://www.vaio.sony.co.jp/Info/Cm/index.html http://www.mastercard.com/jp/about/cm/ http://www.i-love-epson.co.jp/park/cm/ http://www.nttdocomo.co.jp/museum/cm_gallery/home.html |
| Fri 25 Apr | anomolous coward | Slashdot story on the Honda ad, with discussion, links to articles, and the ad itself: http://slashdot.org/article.pl?sid=03/04/14/1146230&tid=129 |
| Sun 27 Apr | Albert D. Kallal | Wow, nice links! Well, I guess a lot of stuff is out there...just got look more! |
| Sun 27 Apr | www.marktaw.com | Commercials used to be minutes long, and now commercials are 30 or 15 seconds, but shown more often. Studies that were conducted a while back showed that the most important thing to getting someone to pick one item off the shelf v. a competitor was brand recognition. If I don't recognize a brand I'm not going to use it. By the same token, newer, fresher, ads keep products in people's minds, and ensure they keep thinking of them as new, fresh products. If there were a spate of Coke commercials from the 80's I would probably start thinking of Coke as an antiquated product from a bygone era. Our memories of great ads also may be better than the ads themselves. 'Geeze, the CocaCola corporation thought Michael Jackson was cool. What morons.' 'I never realized how low budget that 'Where's the Beef' commercial was.' |
| Software proposals | Thu 24 Apr | #If |
| Anyone know of a link to a software project proposal template? Thanks |
| Thu 24 Apr | flamebait sr. | 1) Design software 2) Make Software 3) ..... 4) MAKE MONEY! |
| Thu 24 Apr | #If | What was #4 again? |
| Thu 24 Apr | Dennis Forbes | More like #1 Think up grand software plan #2 Put up webpage #3 ... #4 Profit! |
| Thu 24 Apr | one programmer's opinion | What a lot of consulting firms do is cut and paste words from one proposal to another. You might find a few by searching various state government run web sites. Another alternative is to buy a software product such as: Proposal Kit Pro http://www.proposalkit.com/htm/propkit.htm |
| Thu 24 Apr | #If | Thanks OPO! |
| Thu 24 Apr | Prakash S | Check Steve McConnel's website. |
| Thu 24 Apr | Matthew Lock | A little trick you can do in Google is search for the name of the file you would like to find. Try searching for this: software project proposal.doc You often get a whole heap of word documents. Try PDFs as well with: software project proposal.pdf |
| Sun 27 Apr | #If | Matthew... Outstanding, Thanks! |
| Worst tools I ever did a successful project with | Thu 24 Apr | Mike |
| What is the worst toolset you have made due with on a project that turned out successful |
| Thu 24 Apr | | Visual C++ & MFCs. |
| Thu 24 Apr | Dan Maas | Not a software engineering project but: Adobe Premiere. The worst video editing software the world has ever seen. |
| Thu 24 Apr | Better than being unemployed... | Bespoke 8 year old library (written in house by long departed developers) that wasn't really designed for the scale and performance I was using it for. |
| Thu 24 Apr | Bella | Oracle Forms/Developer-2000 was the biggest POS I've ever used. The textbased Oracle*Forms was groundbreaking in it's time, but Dev2000 was a joke by the time it was released (Powerbuilder and VB were already entrenched by 1997) |
| Thu 24 Apr | Mitch & Murray (from downtown) | Symantec C++. What a POS. |
| Thu 24 Apr | Gregor Brandt | Rational Rose UML Toolset. Don't know what its like now, but a few years ago, what a piece of junk! Horrible UI, inconsistent menus, crashes.... |
| Thu 24 Apr | Matt Christensen | Win NT 4 sp 3, MS SQL 6.5, + ASP. I swore of Microsoft products for 2 years after that debacle. 6.5 had a bug where my identity columns would corrupt and start throwing key clash exceptions; IIS would crash way too frequently, generally taking out the kernel with it. Lest you think I am an MS hater, W2K and MS SQL 2000 are both favored products of mine now... Next would be mod_perl + MySQL. Modperl wasn't too bad, but MySQL at the time had a bug that caused me to have to bring down the db, manually re-index and repair the tables every night. Not fun. |
| Thu 24 Apr | KenB | Interwoven Teamsite - utter crap... |
| Thu 24 Apr | Ben Thompson | Interworld. Thankfully I was lucky enough not to have to go near it but its template language was a thing to behold. No subroutines or case statements leads to hideous code. Early circa 1996-7 ASP was equally fun (more due to the pain of writing ActiveX controls than anything else mind). |
| Thu 24 Apr | Poor Ideafix Developer | Ideafix for Windows. You don't know it? Thank your gods for that, because it's the worst piece of commercial software you can ever see. |
| Fri 25 Apr | Just me (Sir to you) | PhP, no contest. |
| Fri 25 Apr | Better than being unemployed... | Another one to add : Windows NT or 2000 dev PC accessing a SourceSafe database running on a Novell server. Can you all say 'S L O W'. We badgered IT for what seemed like years to get the server changed. People took tea breaks whenever they checked out a file. Eventually they switched to a 2000 server and productivity shot up. |
| Fri 25 Apr | Dale | One time I had to work with this guy named Dale. Dale and I were able to complete the development work on time, but MAN was that guy ever a huge tool. Huge. |
| Fri 25 Apr | Chris Tavares | the pSOS real-time OS. Man, what a pain. The docs were confusing and often wrong, it took us several weeks just to configure the darn thing to download, it supposedly came with a debugger, but the debugger only worked when the board was running (which is the only time I don't need a debugger), and so on and so on. |
| Fri 25 Apr | Beka Pantone | Well, in my experience: PHP wins hands down as the most frustrating 'tool' I have ever used. |
| Sun 27 Apr | Ethan Herdrick | FutureTense application server. No question. The very name of this thing makes me - and the rest of the alumni of the project that used it - laugh out out loud, then cringe. The CTO of the company that produced this horror actually came to our offices to try to defend it. The classic quote from him was 'If we allowed you to do addition or subtraction in our language, that would be slippery slope to who knows what!' I guess you had to be there. But trust me, it made the things you other guys are talking about look like miracle tools. |
| Immunogenetic Computing | Sat 26 Apr | Artist |
| Hi, If anyone is aware of Immunogenetic Computing? I search but not avail. Thanks, Artist |
| Sat 26 Apr | Simon Lucy | http://www.cs.kent.ac.uk/people/staff/jt6/aislinks.html Look for Immunological Computation or Artificial Immune Systems. |
| will struts save me time? | Fri 25 Apr | choppy |
| I have a contract coming up to build a series of wizard-like input forms for a java web application. Most fields will need some amount of validation but nothing complicated. I have bad memories of a previous project where I ended up writing tons of servlet code for what I thought was going to be simple form handling. Im trying to find the best way to do this writing the least amount of code. Will the DynaActionForm in struts help me out in this regard? |
| Sat 26 Apr | Daniel Shchyokin | NO |
| Sat 26 Apr | choppy | is there anything that will? |
| Sat 26 Apr | Matthew Lock | I found that writing a kind of mini-language helped the most. That language let me define online forms, and then would generate the HTML and validation code needed. Once the initial effort of passing the mini-language was done I could knock up a very complicated form in minutes. |
| Sat 26 Apr | Robert Chevallier | The Struts taglib can help you for quick easy field syntax validation. Combined with the power of ActionForm (method validate), you can do whatever you want. Only problem, it takes some times IMHO to really understand well Struts, so you may not gain a lot of time for your 1st use. Struts is for the time being the de facto standard for java web MVC application. I do recommend it. |
| Sat 26 Apr | Konrad | I've got about 6 years of Java experience - I wouldn't say I know it perfectly but I'll have a good idea about what to do in most situations. After a nasty time coding a JSP+servlets web front end to an application, I decided for the next one I would use struts... I only had two - four hours a day, and it took me about six weeks to get to a 'business as usual' level of struts knowledge where I'd know how to do something if asked. It was perhaps compounded by having to learn an object-relational library at the same time, but it did take me a lot of time to get up to speed with struts and how to achieve things that I could easily have achieved in plain servlets and JSP - there are a few quirks. The flipside is that once you are established with it, it is versatile enough to do the job, combined with Tiles, you've got a very effective platform for creating web applications and I now prefer to use it on small jobs as well. In short, if you are under time constraints, I'd recommend that you don't go ahead with using struts unless you can afford to lose two weeks, but that's only based on my experience. I'm not sure whether I'm a slow learner or not, based on past experience I'd say no, but your mileage may vary. |
| Sat 26 Apr | /good-advice/1624/use-struts.do | You can use Struts Validator component and write all your validation rules into XML configuration file. Since you wouldn't need custom ActionForm classes for validation, DynaActionForm is the only ActionForm you'd need. |
| Signing software packages | Fri 25 Apr | Beka Pantone |
| Im writing a little pet project that involves automated software updates. To determine the packages authenticity these kind of services normally sign the packages cryptographycally. This is the first time Im confronted with PKI cryptography in a software project so Im a little confused. Heres what I know so far, please correct me if Im wrong. * Package signing can be done with PGP (or gnupg). * I understand that the server delivering the software packages will have its own set of keys (public and private). * The clients need the servers public key to verify the packages, however embedding the servers public key in the client software is a bad idea. Should the servers private key be compromised the servers key pair must change and clients should be aknowledged. Clients should then get the servers new public key. The hairy bit for me is, how do clients know the new key is safe? A malicious user could be spoofing the service providing a new set of keys and properly signed software that will pass the verification rules of the client. I understand there are companies like Verisign or Thawte that will issue digital certificates that will address the problem of autheticity of the keys. But is this really necessary? I mean, I would like to be able to change the keys at any time without discontinuing service and not have to put the service offline untill some company issues a new certificate. As an example, many apt-rpm repositories contain signed packages how does the client know package PackA comes from Repository#1 instead of EvilRepository#2 posing as Repository#1? Any help is appreciated. |
| Fri 25 Apr | Dan Maas | I think end-users are generally supposed to get the public keys from a secondary source they trust, like your company's website. e.g. http://www.softwareco.com/public_key. Of course it's possible for an adversary to take over your domain or spoof the HTTP traffic, but that's quite a bit harder to do than just embedding a fake public key in the software download itself. As I understand it, companies like Verisign exist to provide an extra layer of security, to verify that the aforementioned adversary has not actually spoofed the connection you use to get the public key. |
| Sat 26 Apr | Nice | go ahead and embed the public key in your app. Will people really be trying to factorise it or whatnot? On the other hand, keep your private key on a computer that is air-gapped from everything. |
| So everyone has read E-Myth | Fri 25 Apr | Daniel Shchyokin |
| Any other start your own business type books people like? |
| Fri 25 Apr | | E-Myth was OK, but really a summary paragraph would have been sufficient. It seemed very repetitive to me. Aside from that, what you need are business forms, tax guides, accounting/bookkeeping books, etc.... Not very glamorous stuff. |
| Sat 26 Apr | Joseph Grace | There are an untold number of microbusiness books out there. The easiest recommendation(s) are those from Nolo Press. Nolo specializes in providing legal advice so you do not have to go to lawyers for basic stuff. The books are written by lawyers and they write great books. www.nolo.com Their books are inexpensive, and many are even at 30% off sale for now from Nolo direct. Some faves I've just purchased: 'Legal Guide for Starting and Running a Small Business' Steingold (7th edition, Nolo) 'Legal Forms for Starting and Running a Small Business' Steingold (2nd edition, Nolo) complement to 'Legal Guide...', above 'Tax Savvy for Small Business' Daily (6.2 edition, Nolo) 'This plain-English guide will show you how to make the most of your tax deductions.' --- Business Week 'Marketing without Advertising' Phillips and Rasberry (4th edition, Nolo) 'High-impact, low-cost marketing strategies...' You may also consider any of the 'how to incorporate', 'writing a business plan', 'copyright and trademark', or other legal books they have. Great books, great company. There are a multitude of other great books on writing business plans, managing employees, providing a positive customer experience, bookkeeping, selling, etc.. You can check Amazon for reviews and for links to other 'recommended' books. Find a book *you* especially like, and follow the links to other books by people who purchased that book. The trail is virtually endless. Some starting points: 'Selling the Invisible' Harry Beckwith (Warner Business Books) Bite-sized micro business marketing tips (first of 3 books) Recommended by a friend 'The Tipping Point: How Little Things Can Make a Big Difference' Malcolm Gladwell (Little, Brown publishing) Meta-business book or 'working *on* the business' in E-Myth terms There are plenty of great books. You probably do not want to read all of them though. So, I suggest three caveats: 1. Try to stick to books which apply directly to microbusiness (exception: 'The Tipping Point', above). The bulk of best sellers are for corporate america, not microbusiness. Most great books are irrelevant to your situation, and your situation will be very fluid as a startup. Those books have little (if any application) when running a small business. For example, a 'How to become a great CEO' is useless for a microbusiness as it targets corporate environment issues; it is easy to get carried away from Amazon reviews. So take five-star reviews with a grain of salt until you see the book and ensure that it's not just megacorp-centric. Plus, browse the books of interest to ensure they speak to *you*. 'Nuf said. 2. Only buy book(s) that save you money (e.g., Nolo legal), have direct business relevance (e.g., customer service, or generating sales leads), or for which you have a craving (i.e., use your craving to learn) *now*. Try to read at least portions of the book(s) right away to satisfy your craving or need. Browse your local Barnes&Noble (for example) to determine which books suit your tastes and needs, and to scope other books. Try to pare down your purchases to just those book(s) you care about and shall read *now*. 3. Not all great books are meant to be read by you. Another filter for books is length. Short books are often better than longer books. Gauge how important a topic is for *you* right now. If it's important but secondary, consider purchasing the shorter of two great books on the topic. You'll probably be better off reading the whole of the shorter book than a fraction of the longer book. Your satisfaction will be greater as well (and you can always save the name of the longer book for later consumption). IOW, match the book length and your time-commitment to the value of the topic to you. Incidental topics suggest shorter, bite-sized books. A good example would be the Beckwith (bite-sized marketing advice) versus the Nolo's 'Marketing without Advertising' (structured marketing advice). Each books serves a different purpose; they're both great. For a brief read, go Beckwith; for a full planning guide, go Nolo. They both have their purpose (I have both), but when in doubt, go lightweight (shorter, more topical). The heavyweight stuff will still be there, and in the meantime you'll be better prepared for the heavyweight stuff on your next visit to the topic. That's more than enough for now! |
| Justifying monitors | Fri 25 Apr | Philo |
| Why is it so very, very hard to justify to management that large monitors are a good idea? - They help me do my work - They make me more efficient - (IMHO) they aid retention - (IMHO) they help recruitment Regarding the last one - if I was interviewing at a place and saw that all the developers had dual head 21CRT & 17 LCD, that place would go to the top of my list (not just the nice workstation, but it gives the appearance that management takes development seriously). A 21 monitor costs less than $500. Thats 3-5 hours of a developers time, to get the benefits listed above for 3-5 YEARS. This should be a no-brainer. Philo |
| Fri 25 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | Agreed 100%. Joel's test: 'Do you use the best tools money can buy?' That includes high quality monitors. I don't compromise my vision for any job. |
| Fri 25 Apr | mark | I actually have a 21' monitor, it was given to me when I first started here more than 2 years ago. And the Dell Laptop they just gave me a few weeks ago has dual monitor support, so I guess I'm currently living in display Nirvana. |
| Fri 25 Apr | SteveM | I switched from a very good 19' CRT to a dual-head 17' LCD system. Anyone who tries to make me change back is going out the window! I find it incredibely frustrating when I have to use someone else's PC and they only have a single monitor, no matter how large. How the hell do you use two programs at the same time on a single screen? ;-) |
| Fri 25 Apr | Joel Spolsky | Here's how to justify it. LCD monitors last 7 years, easily. A second 1280x1024 LCD monitor is about $500. That's $71 a year or $0.20/day. A roll of charmin toilet paper is about $0.75. If you go to the bathroom once a day, and here I'm making some assumptions, you consume about 1/3rd a roll of toilet paper, for a cost of about $0.25 a day. So you are literally asking them to spend less on this second monitor than they spend on your toilet paper. |
| Fri 25 Apr | Peter Ibbotson | Dunno how to justify it (except using Joels method), however we have matrox Dual head cards because they're the only ones that give Win2K/XP TWO video cards in setup. This is important as some of the other video cards treat the screen as one big (but wide e.g. 2048x768) system. This means you can't have different resolutions. Here I have a little 15' LCD running at 1024x768 plus a 21' monitor running at 1280x1024. (the resolution actually matches quite well) The other advantage is that dialog boxes pop up in the middle of the screen rather than half way between the two monitors. Several of the card vendors ship programs which are supposed to fix this problem but none of them actually did (or were very buggy) Note I did most of the research on this during the WinXP beta program with the ATI dual head card so this may have changed by now. Checking the matrox website reveals that the G550 Dual head is still available but the different resolution stuff is broken on the Dual DVI version in DVI mode. |
| Fri 25 Apr | Lou | Don't forget power savings and time saved during reorgs. I've seen references to studies (yikes, 3rd generation knowledge) stating that over 5+ years LCD panels pay for themselves over similarly sized CRTs. |
| Fri 25 Apr | Stephen Jones | The irritating thing is that you can probably get a new processor and motherboard, even if all you do is word processing or data entry, because that does not sound like luxury, but two monitors - no way. Do a time and motion study showing how much time is wasted opening and closing windows. Throw in some made-up figures about increased chance of RSI as well and you may suceed. |
| Fri 25 Apr | Stephen Jones | ----' If you go to the bathroom once a day, and here I'm making some assumptions, you consume about 1/3rd a roll of toilet paper'--- Pretty frightening assumptions if you ask me! I'd try and choose another equally trivial example. I can think of plenty of companies that would simply downgrade the quality of toilet paper, and then you'd be spending all your time making up justifications for a more comfortable office chair. |
| Fri 25 Apr | Mitch & Murray (from downtown) | The new ATI cards allow you to specify the resolution for each monitor. I have an ATI Radeon 9000 Pro ($100) and am driving a 20' LCD panel at 1600x1200 via the DVI output, and driving an 18' LCD panel at 1280x1024 using the analog SVGA output. Two monitors, one card, configure each output separately. Works flawlessly. Pretty neat. Sounding a bit less professional, I can also report you can run Flight Simulator 2002 in _amazing_ ways with this setup. |
| Fri 25 Apr | Dan Brown | Here is the argument I use (I've equipped everyone here with at least one 20' 1600x1200 LCD, and most people have a second LCD as well. The company is mine, so it is my money I'm spending): The advantage of large displays is that they allow you to see at least two pages side-by-side, if not more. Most programming these days involves using existing components (such as classes) or creating components to be used in larger programs. That means that as a programmer, you almost always have your code at at least one interface to deal with at the same time. Having a large screen allows both your code and associated interfaces to be on the screen at the same time. In my experience, this dramatically reduces bugs. There is a simpler way to look at this: the more of your program you can see at one time, the less likely you are to make mistakes coupling pieces of the program. The argument for multiple monitors is a little different, since you generally have to break your concentration a little to move from one to the other. I think the second monitor is useful for reference material and other applications. For example, almost all reference documentation these days is provided only in computer-readable form. If you want to have that reference material easily available during programming, a second monitor helps much more than switching applications on one monitor. For example, if you are working on hardware drivers, it is very convenient to have the PDF files containing the datasheets and user manuals for the chips you're working with displayed at the same time you are writing the code. The cost of the monitors is quite reasonable. A decent 20' LCD screen can be bought for about the cost of a week of programming. |
| Fri 25 Apr | Mitch & Murray (from downtown) | 'A decent 20' LCD screen can be bought for about the cost of a week of programming' I've seen the excellent Dell PF2000 (the twentyincher I have) for as low as $850 when Dell runs one of their 20% off accessory sale + online coupon fandango blowouts. For $850, buying that monitor is a no-brainer. 20', 1600x1200 native resolution, DVI, analog, S-Video, and NTSC video inputs, PIP. |
| Fri 25 Apr | David | Someone explain this dual monitor thing to me, especially the mix between CRT and LCD. Is that just because two CRTs are unwieldy? Are LCDs easier on your eyes when you look at them all day? If you have a smaller LCD, like say 15', do you just use that for your documention (class help, google searches, etc) while you program? I doubt my struggling company would go for it, but I'm curious nonetheless. |
| Fri 25 Apr | Mitch & Murray (from downtown) | The LCD panels are immensely easier on your eyes. Considering you stare at the thing all day this is a pretty major update to your work environment. Right up there with a decent chair. Chair, monitor, keyboard, everything should fit you, cause that is what you are screwing with all day. I wonder how many folks here are, like me, hardcore trackball converts? The common LCD - CRT combination arises after the lucky developer gets an LCD panel and then muses 'What in the hell should I do with this old CRT?' After a few moments the magic word 'Presto' is spoken and connections for the second display (in this case the old CRT) are established. |
| Fri 25 Apr | Philo | I'm CRT/CRT but hope to move to CRT/LCD. Reasoning: I run a 21' CRT at 18x14. No reasonably priced LCD runs that high (I wasn't aware of any LCD's that ran at 16x12 for less than $1500 until just now). My work area is two folding tables butting together, and the LCD is in the corner, so it's using space I wouldn't use anyway. To my left is a 17' CRT at 1280x1024. I have six inches of usable desk in front of it, which isn't really usable. I hope to replace it with an LCD set back so I have room to set a book in front of it. I generally have Enterprise Manager open on the left screen, with Visual Studio open on the 21' main screen. Sometimes I may have a web page, UltraEdit with a few text files, or XMLSpy open to the left. Terminal Services goes over there on occasion as well. But in essence the left screen is for reference material, the main screen is for work. My dream is three screens/two boxes: LCD CRT LCD \ | / \ | / \ KVM / \ / \ / \ / \ / CPU1 CPU2 (It's a mad, mad, mad, mad setup) Philo |
| Fri 25 Apr | Andrew Hurst | Similar to Philo's dream setup, but not all the way. I have 2 21' monitors on my desk. The second monitor is connected to a KVM that connects to two other computers. Works out well. This discussion has made me think more about LCD's though... |
| Fri 25 Apr | SteveM | The main advantage of a dual monitor (for me) is that it makes far easier to debug drawing code. Stepping through drawing code is _so_ much harder with one screen, since anything that pops up over your program will affect the code path. There are other ways of course - remote debugging on another machine or the wonderful VMWare, for example - but the second monitor is probably no more expensive. Oh - and it makes it much harder for my MD to spot my web-surfing when he sticks his head round the door! (Sorry :-) |
| Fri 25 Apr | Dimitri. | I would never use a CRT again. LCD's are much easier with your eyes, so it's no question at all. I simply refuse to work with CRTs. And when the price difference is so small (from a company's point of view, anyway) there is simply nothing to discuss. A good way to go for a dual-LCD would be to ask first for the small one (15') as they cost around 300$ now and it would pass very easily. Wait two motnhs, then ask for the big one. That may work. Or not. If you think that they'll say 'wait, we just bought you one, you want another?' then you can have the big one first, and when you ask for the second one, they'll be relieved at how cheap it is. I have a dual-head ATI Radeon (VE) and it sucks more than anything that ever sucked before. After installing drivers a few times, you finally can enable the second monitor, but only at the same resolution. Finally, I somehow managed to have different resolutions, but with the frequency at 60Hz (frequency is less important on LCDs than on CRTs but still I would like to have it at at least 70). When I finally had it running with both I realized I must disable one of them, otherwise Day of Defeat (it's a programming environmens, and yes, it has source control) would crash like a stone through a lighthouse. So now I use both but disable one sometimes, which works. Regarding more than two displays: didn't Matrox have some form of 'four-head' card? One that costed a fortune? |
| Fri 25 Apr | Philo | The last time I worked on either 15' or 1024x768 was 1993. I don't see any point in jumping back ten years in time. Philo |
| Fri 25 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | In the consumer space, no, nobody has ever had a quad-head card that I've heard of. Matrox has a tri-head card that they demo'd as a primarily gamer's card (180 degree immersion which can be very nice for flight sims and the like). Nvidia definitely has PCI versions of some of its select cards, so a quad head should be as simple as an AGP and a PCI card. Where the hell are you going to put all the displays? :-p Three I can get, four I don't (unless you have a special 2x2 enclosure). |
| Sat 26 Apr | David Fischer | Dimitri wrote that he has ,'a dual-head ATI Radeon (VE) and it sucks .... After installing drivers a few times, you finally can enable the second monitor, but only at the same resolution.' In general, that's not correct. At home I've got the Radeon VE driving a 17' monitor at 1600x1200 and a 15' monitor at 1024x768. The driver install wasn't difficult and I've been able to run them at different refresh rates. The Radeon drivers have their quirks, but Dimitri's problems are not universal. Now, if I can get my work to upgrade my computer and go to a dual-head support for my optical design work (not just for programmers, you know). |
| Sat 26 Apr | Dave | If I wanted to upgrade our development machines that currently have one AGP card each, can I just add a 2nd video card and configure dual monitor support in Windows XP? I believe it supports this, but haven't tried it out myself... |
| Sat 26 Apr | Brad Wilson (dotnetguy.techieswithcats.com) | Yes, you can do it that way. |
| Linux != future | Fri 25 Apr | sedwo |
| Ive been thinking about this for a short while and would like other peoples viewpoint. Constructive criticism please. Linux has been created by hackers, for hackers. And I doubt it will ever truly replace the Windows/Mac GUI systems because the hackers wont let it. Because to reach that state of mass acceptance, you have to think and develop for those people. And the last thing most Linux developers want to do is take the time and energy to creating something any idiot can point and click at. And especially when their not getting paid for it. Windows on the other hand *does* have that. It will *try* to work with the user. And while its not perfect, its a far cry from what Linux does; which in most cases is *nothing*. You want help, go get it yourself. It also seems that some of these open source developers pride gets bruised everytime a user makes any feature comment on Linux looking/acting like Windows. Immediately the look and feel gets changed to make it *clearly* known to the user that *THIS IS NOT WINDOWS*. Deal with it. Now the Linux community knows this, but the mentality is, were hacks. We need something to work just good enough for ourselves. And if youre not a hacker, then you shouldnt be using this. Thats fair. And thats the stigma that youre stuck with, so stop complaining about why the world hasnt switched to Linux. Oh and that free thing. We all know its really not free. The total cost of ownership for a Linux box is far more then just the OS. Money turns the world, and free ends up costing more then not. I cant predict the far future, and while Linux has its place; IMHO that is where it will always be. As scary as it seems, it truly is a Microsoft world. And most likely will be for a long time. |
| Fri 25 Apr | Mitch & Murray (from downtown) | The main contribution of Linux seems to be giving people the opportunity to dump a Unix based proprietary hardware/OS solution (Sun, SGI, etc) and run a generic Unix clone on cheap x86 hardware. One of these days this combination will be a death blow to Sun, at least on the desktop and small/medium server side. I have been using Linux on a daily basis the last few months for a Unix based project. Running RH8, all I can say is the desktop client is still a slow, clunky joke. I don't give a shit if the code is 'free', give me something that works as well as Win2K or XP. So much stuff is quirky or just flat-out broken it is amazing that so many Linux zealots beat their chests over how good this stuff is supposed to be. Replace Windows? Never going to happen on the desktop, will probably never happen in Microsoft-centric data centers. But Linux will, over time, replace the other Unix flavors, and the Unix hardware vendors with their proprietary Unix operating systems, are doomed. Your mileage may vary, void where prohibited. |
| Fri 25 Apr | Just me (Sir to you) | Hey, I know the whole cryonics thing was cool in the late 70's and early 80's, and by God who would have believed it would really work. It was even funny at first, all these IT defrosties just going about their business as if the world after 84 had never happened. You know, working on the most bombastic kludge rewrites of 70's UNIX implementations and the like. But then some politico radicals got involved (got to love that neo-Orwellian newspeak) with smooth marketing pulling some sort of extra slippery Buck Rogers on some Finnish dude and slick branding with some 'I'm maladapted but cute' animal for the puberty hormone swigging crowd, and it got less and less funny. Now they want all of us that learned how to stop dragging our digital knuckles during the 90's to stoop down again? Count me out for the ride pal. (Hey, this feels like I am repeating myself) |
| Fri 25 Apr | John Topley | LMAO! |
| Fri 25 Apr | Colin MacDonald | The Lindows folks (http://www.lindows.com/) are actually heading that way. Their systems are big on the flashy GUI, quick links for browser, mail, and IM, easy upgrade tools; and short on geek features (no apache, ssh, or most command-line tools). They're really geared for average users. But it's still only a second-rate, wannabe Windows - why would anyone settle for less? There are actually a couple reasons. First, it still has lower hard-cash costs. This is a factor when you get outside the first world. India and China together have, what, about 10 times the US population? And a few hundred bucks for the Windows OS + Office isn't chump change there. The other reason is that nobody really owns open source code. Specifically, it's not owned by a large American company. Even in Europe, where cost isn't so much the issue, people (particularly in government) are starting to think that it's not such a good thing that Microsoft has a death-grip on all their computer systems. So, Linux may not be your future, or America's future, but it may well give Microsoft a run for its money on the bigger playing field. |
| Fri 25 Apr | Better than being unemployed... | I don't think any sane, unbiased, Linux afficianado would disagree that if you want to develop desktop applications, you've got to use Windows. However, if you're in the business of making servers and server-side solutions, where there's no GUI involved, then Linux is definitely worth considering, especially where you compare performance and reliability. For instance, compare and contrast Exchange server to a UNIX based mail server : http://www.catb.org/~esr/fetchmail/fetchmail-FAQ.html#G8 (which claims that Microsoft use UNIX servers for Hotmail). I think most people realise this, and are concentrating on getting Linux into the server market. This probably explains why not much work has been done on the GUI, as nobody's concentrating on that as much. Personally, I don't really mind what operating system I'm using as long as it gets the job done. I use Linux for server side processing, and Windows for running client apps, and that works fine... |
| Fri 25 Apr | Scotty | I have no hope for Linux as a desktop operating system within the next few years. However, we run Linux and OpenBSD on all of our servers, and have been for many years now. The server is the past, present, and future of Linux. |
| Fri 25 Apr | Just me (Sir to you) | 'which claims that Microsoft use UNIX servers for Hotmail'
Yes, it even claims it runs on Solaris (which it never did), and refers in its documentation to 'M$'. Bet their programmers are truly l33t too.
They're a mail client outfit and they refuse to adapt their stuff to work with Exchange? |
| Fri 25 Apr | Stephen Jones | Linux will run at the bottom end of the market, since you can surf the web, write letters, do spreadsheets and indeed most things 90% of home users do on a computer with Linux. If the hardware's compatible Linux will pass the grandmother test as well or better than Windows. Indeed for most tasks you won't even be aware of what OS you are using. Linux is already taking up much of the mobile computing market; the competition is Symbian, not Pocket PC, which is the preserve of companies 100% MS, or consultants working with those companies. Linux will also get government stuff outside of the US, as where computers are only used for one application there is no need to waste a fortune on MS licenses, and the present almost universal fear of US foreign policy will act as another reason for avoiding MS dependency. So Linux/Unix will be taking the server end, and the bottom end of the consumer/office market. The application bar to entry will allow MS to keep the middle part of the market. And if it starts to lose out there it will simply drop prices. |
| Fri 25 Apr | Philo | WebTV ran on Unix until at least 2000. When MS bought WebTV they issued a directive that all Microsoft properties would migrate to NT (because they were tired of getting bad press about Hotmail running on Unix). WebTV prepared an estimate of the hardware and admin costs of migrating to NT and submitted it. They immediately got special dispensation to keep running on Unix. My source is a friend who was a founding senior tech at WebTV. Philo |
| Fri 25 Apr | Mike Gamerland | Making my living in both arenas (Windows and Unix/Linux) I have a slightly different view. I know the JOS tends to be MS centric but I believe you may be underestimating or merely uninformed on some of the particulars. While I can make a comfortable living on Windows, I also see expanding opportunity in Linux, so I generally recommend to my developers that they expand their horizons. Linux was created by hackers. - Using the true form of hacker this is fairly accurate. A few folks got together and created the kernel of what we call Linux. However, the same would be said for DOS which later became the underpinnings of Windows. Today, I merely look at hackers as people developing code without corporate funding. To remove the negative connotation of 'hacker' as someone doing something illegal or unjust, Open Source is now a more popular use of the process. Linux and GUI - Interestingly, Linux (okay the BSD derivative) is Apple today. So, is that a win? Let's say 'no' because apple has only 4%(?) of the desktop. However, to the point of usability, Linux was not a concept that started to work at replacing windows. Therefore its original audience was not 'any idiot' as you put it. As for windows working for you, in that r |