Ditto for me: I'd love to see some more examples of large LAMP systems under test. This is one case where testing the right things, and the right way, seems non trivial.
The other time I've had a really hard time figuring out the right way to do my testing was working on an aggressively multithreaded and distributed server application.
Thanks for the input; I'll give it a look. But what I'm really after is larger scale real world systems. I've read quite a few books that walk you through the process, and I've written my share of tests on various projects, but there's nothing like full scale examples to point out the pitfalls.
Thanks for the pointer! I hadn't thought about either (S)Cons or Aegis in quite awhile; it's good to be reminded. I will definitely spend some time reviewing the tests that come with SCons, and try using it for some of my projects. Nothing like build scripts in python to make you appreciate how painful make files can be;)
It's been interesting how most of the examples people have come up with are Java and Python; seems to be a language culture kind of thing.
Mostly I write C++, Java, PHP and Perl. But I didn't want to limit the question by language, since I'm interested in pretty much any project using the methodology regardless of implementation platform. Thanks for the Struts Applications pointer!
I agree in spirit, gui file selectors drive me crazy 'cause it's so much harder to quickly get where I want to go. But rather than:
by extention, have "/u*/l*/b*/mozilla" typed into the text area find/usr/local/bin/mozilla if that's the only match.
I want filename completion instead. You get immediate feedbase (if/usr doesn't exist, it doesn't complete it, so you don't waste time doing/u*/l*/b*/mozilla only to have it fail...) I know that some file picker dialog boxes are smart, but I never remember which ones, and so I'm always tabbing myself to the next field. Grrrr.
Similarly, I know there are some file picker dialogs now that do correctly interpret../../whatever.txt, but again, I'm always forgetting which ones;)
I'm with you on your third example. I too would like to see wildcards work for what they're good at; selecting multiple files quickly.
This is the first color image of Mars taken by the panoramic camera on the Mars Exploration Rover Spirit. It is the highest resolution image ever taken on the surface of another planet.
There's clear compositing artifacts in the image (where the subpieces don't stitch together smoothly), so I got to wondering: what's the previous record holder? And was it a single image or also a composite?
For me at least, quitting was the easy part; staying quit is hell. I've been a heavy coffee drinker since early high school (I'm now 27) and absolutely love the stuff. It smells good, tastes better, and the feel of caffeine entering my bloodstream is heaven. But I get fierce headaches if I go without, some other compound in coffee gives me nasty body odor, and I hate the feeling of being so dependent on a substance.
So a couple of years ago I decided to quit. I personally had to go cold turkey; tapering off was too hard to contemplate. And yes, it sucked for a week; bad headaches, tired, cranky, etc.. I second the good advice already given to sleep alot and drink lots of water. After a week though, the physical symptoms are pretty much over.
Then, for me at any rate, the craving really set in. I sort of figured that after awhile it would just be out there, something I didn't drink anymore and used to like. After years of not drinking coffee I lust after it more than ever. Walking past the coffee aisle in the grocery store is torture. Hanging our with friends in a coffee shop makes me want to chew my nails off. All I can say is, I'm glad I never got going on something like cigarettes or heroin.
So my advice? Think hard before quitting. Do you really want to? Are you willing to go through life always wanting it and not doing it? For me, for some perverse reason, the answer is yes. But if it isn't for you, why put yourself through the initial pain of withdrawal?
He also touches on, but does not explore with a self critical eye, the limitations imposed by not having source code. The dependence of Windows programmers on Microsoft APIs provides too many limitations, and increases the likelyhood of unforseen interactions that cause bugs. He whitewashes these issues by simply focusing on the size of the Windows desktop deployments vs. *nix.
Interestingly, I think this part of your comment mostly proves his point;) When programming in a UN*X culture (ie. mostly for the benefit of 'programmers'), the lack of source is a near fatal flaw for the stability ('goodness' of code) reasons you point out. But if your cultural bent is to be worrying about users, then it's patently absurd to develop for a platform with <10% of the entire user base. You're never going to be seen by enough of the users!
Like to program,
are properly trained and schooled,
who are paid enough,
and are given enough time to do the job right.
Seems to me that it's the are properly trained and schooled part that is under discussion. What methodological tools (ie. XP, etc.) can programmers use to best do the job right?
I don't understand any of your examples. Nothing stops you from running KDevelop under Gnome, nor running whatever web browser you seem to feel is tied to Gnome under KDE. If fact, I don't run either (if run is taken to mean use them for my desktop environment) although I use many apps written for both. As long as you have the libraries installed, it's all good.
The GNOME/KDE choice is annoying. Honestly I don't care which one goes away, I just wish one of them would.
How is this different from "The Windows/Linux choice is annoying. Honestly, I don't care which one goes away, I just wish one of them would."
This project looks interesting, and it reminded me of something I found looking through the LinuxFund project proprosals. The ganesha's project people are working on developing a school LAN using relatively old PCs and various free software. You can vote for the project to receive funding from the cool LinuxFund people at this link.
I can't help but wonder if some technical books might be even harder to get (and correspondingly useful as donations) than x486 class hardware components.
When People magazine does an issue devoted to "what's hot" in fashion, do they interview Jane Doe from Des Moines, Iowa? No.
Agreed. But the quality of analyses should be evaluated on the basis of their goals. Was this supposed to be a "what's hot" story, or a "what are most people buying" story, or a "who makes the best work clothes" story? I could personally care less what's hot in fashion.
A technology company presumably put more informed thought into their choice of
server than an art supply house or whatever.
I would expect this to cleave along different lines. To my mind, companies for whom their computing platform is central will put more informed thought into their choice of server. Printing houses, post offices, polling companies, etc. are certainly as reliant on their computer infrastructure as technology companies (software dev, online sales, etc.) but will almost certainly have very different selection criteria.
Slashdot Mirror
Ok, if you have a particular fondness for lame sexist jokes I can maybe (maybe!) understand +4 funny. But insightful? WTF?
Ditto for me: I'd love to see some more examples of large LAMP systems under test. This is one case where testing the right things, and the right way, seems non trivial.
The other time I've had a really hard time figuring out the right way to do my testing was working on an aggressively multithreaded and distributed server application.
Thanks for the input; I'll give it a look. But what I'm really after is larger scale real world systems. I've read quite a few books that walk you through the process, and I've written my share of tests on various projects, but there's nothing like full scale examples to point out the pitfalls.
Thanks for the pointer! I hadn't thought about either (S)Cons or Aegis in quite awhile; it's good to be reminded. I will definitely spend some time reviewing the tests that come with SCons, and try using it for some of my projects. Nothing like build scripts in python to make you appreciate how painful make files can be ;)
It's been interesting how most of the examples people have come up with are Java and Python; seems to be a language culture kind of thing.
Thanks again.
Mostly I write C++, Java, PHP and Perl. But I didn't want to limit the question by language, since I'm interested in pretty much any project using the methodology regardless of implementation platform. Thanks for the Struts Applications pointer!
Similarly, I know there are some file picker dialogs now that do correctly interpret ../../whatever.txt, but again, I'm always forgetting which ones ;)
I'm with you on your third example. I too would like to see wildcards work for what they're good at; selecting multiple files quickly.
The caption says that
There's clear compositing artifacts in the image (where the subpieces don't stitch together smoothly), so I got to wondering: what's the previous record holder? And was it a single image or also a composite?Any pointers?
So a couple of years ago I decided to quit. I personally had to go cold turkey; tapering off was too hard to contemplate. And yes, it sucked for a week; bad headaches, tired, cranky, etc.. I second the good advice already given to sleep alot and drink lots of water. After a week though, the physical symptoms are pretty much over.
Then, for me at any rate, the craving really set in. I sort of figured that after awhile it would just be out there, something I didn't drink anymore and used to like. After years of not drinking coffee I lust after it more than ever. Walking past the coffee aisle in the grocery store is torture. Hanging our with friends in a coffee shop makes me want to chew my nails off. All I can say is, I'm glad I never got going on something like cigarettes or heroin.
So my advice? Think hard before quitting. Do you really want to? Are you willing to go through life always wanting it and not doing it? For me, for some perverse reason, the answer is yes. But if it isn't for you, why put yourself through the initial pain of withdrawal?
Best of luck.
Cheers,
-emile
This project looks interesting, and it reminded me of something I found looking through the LinuxFund project proprosals. The ganesha's project people are working on developing a school LAN using relatively old PCs and various free software. You can vote for the project to receive funding from the cool LinuxFund people at this link.
I can't help but wonder if some technical books might be even harder to get (and correspondingly useful as donations) than x486 class hardware components.
-emile