... write an atricle riding on the open source buzz. Step 2: Get free publicity for consultancy Step 3: Profit!
It's nothing to do with learning from open source management techniques, it's about employing solid engineering principles. Stating that having good documentation will help a project? Anyone (with a brain) think otherwise?
"Yet it is rare to find a corporate environment where the project team has anything approaching the level of planning, documentation, or review found in successful open source projects."
This certainly doesn't match my experience of corporate projects, but then I may have just been fortunate.
One might equally say "yet it's rare to find an open source project where the project team has anything approaching the level of planning, documentation, or review found in a successful corporate project".
I don't think it's anything to do with whether a project is open source or not.. there are good and bad projects in both the open source arena and the corporate arena. The problem isn't learning from one or the other, it's actually applying the processes and management techniques which are well know, well proved, have been around forever but aren't always employed.
As someone pointed out, Mythical Man Month is a great place to start. I'll take that over 5 pages or general conjecture from any day.
I think here that you've hit upon a fundamental issue - 'expected behaviour' as opposed to 'intended behaviour'.
With no documentation of the existing code, you can write as many tests as you like, but you still can't prove that the code performs as it was originally intended to - you can only prove that the code does what the code does.
Granted that this is of some use for the purpose of creating a regression test suite, but it doesn't alter the fact that your tests can't, by definition, find bugs in the exsiting code, as you've nothing to test against... other than your assumptions...
First of all you need to decide what you want to achieve from your testing. If you're following a methodology such as the 'V'-cycle, then the point of unit testing is to verify that the code correctly implements the design (system testing is where you check that the system implements the requirements).
Many replies here are along the lines of "we don't have any documentation" - well in that case, you can't do truely meaningful testing. You can test what you think the code should do, but that'll always be based on assumptions (and we all know about those...) and it could be argued that this kind of testing is almost pointless - it would be far more productive to concentrate on system testing.
Of course, there's nothing to stop you ammending/clarifying/extending any documentation that you do have while in the unit testing process, as long as you take due care and follow your prescribed process.
Finally, if I may offer some advice - document (and apply QA) your unit tests. In many places there seems to be an attitude that "they're only tests - they don't need good documentation", which in turn creates a maintenance nightmare in itself.
I've got so fed up of spam over the festive season that I finally got off my butt and installed Razor as featured on/. the other day. I've always been kind of against the complete black-hole idea, so Razor was very attractive.
So far I'm quite impressed. Easy to install (a couple of lines in procmailrc) and it's picked up about 50% of the spam I've received so far - importantly it hasn't flagged any legitimate messages as spam. Of course, I reported the other 50%, so that hopefully others won't have to endure them. The nice thing about the systems is that the more people that use it, the more effective it gets. It's not perfect, but in this mean 'ole spam-filled world, it's a good place to start.
With all the comments about using it for faster downloads, etc, etc, people seem to be missing the fact that it'll only really speed your downloads up if you're accessing another site on GEANT. Personally when I was a student, connecting to other academic sites was never particularly slow - but JANET (the UK academic net) doesn't have particularly good peering to transatlantic links (clearly due to the cost).
What GEANT will help make more possible is inter-site co-operation, and apps like high bandwidth video streams. In response to the guy who said it was a waste of money - give it time?
I was basing my asumptions on what I've seen from recent consoles. Maybe it's been limited, but I've played N64, Playstation and Dreamcast games recently, and none of the games (with the notable exception of Gran Turismo) managed to keep me playing for very long. For example, I had 'Echo the Dolphin' on my Mega Drive all those years back, so when I got the oportunity to play the Dreamcast version a couple of days ago, I was quite interested to see what they'd done with the game on a far more powerful platform. Well, it sucked. I'm not going to say that the original was fantastic, but I did spend quite a few hours playing it. The DC version kept me interested for about 1/2 an hour. The graphics were doubtless a lot nicer, but (and it's kinda hard to explain) nothing had been added to the game... it seemed like it had been converted into 3D and re-released just for the sake of it. I'd have been just as happy playing it on my antiquated MD... while the technology had progressed, the gameplay hadn't.
The same strikes me with games such as Soul Calibre, the sequel (?) of Soul Blade on the PS. I know there's a very limited amount of things you can do with a Beat 'em Up, but it seemed to be another case of "we'll chuck a few new characters in and ponce up the graphics. *kerchiiing*".
What I'm trying to get across is the point that are games actually benefiting from more powerful consoles, other than in the graphics department ? Obviously, there's more that can be done with CPU time than polygons.
As for my comment about Goldeneye, I wasn't actually trying to suggest that it was superiour to Q/QII/Q3/Q3A/(insert name here), I was, albiet badly, trying to state that the graphics gave me no incentive to play the game, but that I thought it had good gameplay. (Add to this hours of fun playing in 'Big Head' mode, etc;) I don't hate FPS, but I generally only play them for a short amount of time, and I'm coming back to the 'why shell out for the latest and greatest console if what we've got will do' point.
As for the games in the wing for the X-Box, my knowledge of this area is poor (I'll take a look at the URL once I've finished this), so I'll take your word for it.
Your point about not having to program to a minimum spec is the one I most apreciated, and yeah, I can see this being a good thing. Just so long as MS don't start churning out X-Box SE or X-Box ME =)
Anyway, I'm off to dig out my BBC B for a game of DogFight;-)
Personally, I'm afraid I can't see what's going to be so great about the X-box. I've read the specs, and wow, can it shift a lot of polygons... but common sense suggests that chip makers other than nVida will have bought out darn powerful cards which I can put in my PC by next year.
While people are busy drooling over the number of polys this thing can put out, what about the quality of the games themselves ? Personally, I'm not keen on the Quake genre of games (well, maybe Goldeneye... but that was pure playability, nothing to do with graphical wonders). Could I enjoy such greats as Baldurs Gate and Dialo 2 on an X-Box ?
It strikes me that the X-Box will just become another platform for First Person Shooters and Beat 'em Ups, and regardless of who it's manufactured by, this leads me to the conclusion : no thanks.
From the comments I've read, it seems that people are generally not happy with this move.
Several people have already pointed out that the later RAQs are x86 (K6-2 if I remember correctly) based, and I agree with one of the above posts with regard to seeing Sun's point of view... after all, they're the creators of Solaris, so it's far easier and cheaper for them to support boxes running Solaris than it is for them to support Linux. It's like the MS/Hotmail thing. It was an embarasment that Hotmail ran on BSD, not Windows, and hence MS changed that. Likewise in this case; Sun have spent much mulah buying Cobalt, who's products use, what must be classed as, a competitor to Solaris.
Anyway, on to my point.... first of all, if they do start shipping them with Solaris, is there anything stopping people replacing it with Linux themselves ?
And secondly, why so much anti-Solaris feeling ? I use both Solaris (at work) and Linux (at home). Yes, I see differences, but I don't see what's so bad about Solaris. So, instead of just slaggins Sun off, could people offer some constructive critisism about why this move would be a bad thing ?
Do you think that Bill knows something Steve doesn't ?:) Also, seeing as Bill's been in the 'Opressive Marketing / Global Domination' department for so long, what does he hope to bring to company as 'Chief Software Architect' ?
Slashdot Mirror
... write an atricle riding on the open source buzz.
.. there are good and bad projects in both the open source arena and the corporate arena. The problem isn't learning from one or the other, it's actually applying the processes and management techniques which are well know, well proved, have been around forever but aren't always employed.
Step 2: Get free publicity for consultancy
Step 3: Profit!
It's nothing to do with learning from open source management techniques, it's about employing solid engineering principles. Stating that having good documentation will help a project? Anyone (with a brain) think otherwise?
"Yet it is rare to find a corporate environment where the project team has anything approaching the level of planning, documentation, or review found in successful open source projects."
This certainly doesn't match my experience of corporate projects, but then I may have just been fortunate.
One might equally say "yet it's rare to find an open source project where the project team has anything approaching the level of planning, documentation, or review found in a successful corporate project".
I don't think it's anything to do with whether a project is open source or not
As someone pointed out, Mythical Man Month is a great place to start. I'll take that over 5 pages or general conjecture from any day.
I think here that you've hit upon a fundamental issue - 'expected behaviour' as opposed to 'intended behaviour'.
... other than your assumptions ...
With no documentation of the existing code, you can write as many tests as you like, but you still can't prove that the code performs as it was originally intended to - you can only prove that the code does what the code does.
Granted that this is of some use for the purpose of creating a regression test suite, but it doesn't alter the fact that your tests can't, by definition, find bugs in the exsiting code, as you've nothing to test against
First of all you need to decide what you want to achieve from your testing. If you're following a methodology such as the 'V'-cycle, then the point of unit testing is to verify that the code correctly implements the design (system testing is where you check that the system implements the requirements).
...) and it could be argued that this kind of testing is almost pointless - it would be far more productive to concentrate on system testing.
Many replies here are along the lines of "we don't have any documentation" - well in that case, you can't do truely meaningful testing. You can test what you think the code should do, but that'll always be based on assumptions (and we all know about those
Of course, there's nothing to stop you ammending/clarifying/extending any documentation that you do have while in the unit testing process, as long as you take due care and follow your prescribed process.
Finally, if I may offer some advice - document (and apply QA) your unit tests. In many places there seems to be an attitude that "they're only tests - they don't need good documentation", which in turn creates a maintenance nightmare in itself.
I've got so fed up of spam over the festive season that I finally got off my butt and installed Razor as featured on /. the other day. I've always been kind of against the complete black-hole idea, so Razor was very attractive.
So far I'm quite impressed. Easy to install (a couple of lines in procmailrc) and it's picked up about 50% of the spam I've received so far - importantly it hasn't flagged any legitimate messages as spam. Of course, I reported the other 50%, so that hopefully others won't have to endure them. The nice thing about the systems is that the more people that use it, the more effective it gets. It's not perfect, but in this mean 'ole spam-filled world, it's a good place to start.
With all the comments about using it for faster downloads, etc, etc, people seem to be missing the fact that it'll only really speed your downloads up if you're accessing another site on GEANT. Personally when I was a student, connecting to other academic sites was never particularly slow - but JANET (the UK academic net) doesn't have particularly good peering to transatlantic links (clearly due to the cost).
What GEANT will help make more possible is inter-site co-operation, and apps like high bandwidth video streams. In response to the guy who said it was a waste of money - give it time?
Well, seeing as this : ftp://ftp.debian.org/debian/dists/unstable/main/bi nary-hurd-i386/web/apache_1.3.3- 4.d eb
exists, I assume that some people are running a webserver on HURD ...
I was basing my asumptions on what I've seen from recent consoles. Maybe it's been limited, but I've played N64, Playstation and Dreamcast games recently, and none of the games (with the notable exception of Gran Turismo) managed to keep me playing for very long. For example, I had 'Echo the Dolphin' on my Mega Drive all those years back, so when I got the oportunity to play the Dreamcast version a couple of days ago, I was quite interested to see what they'd done with the game on a far more powerful platform. Well, it sucked. I'm not going to say that the original was fantastic, but I did spend quite a few hours playing it. The DC version kept me interested for about 1/2 an hour. The graphics were doubtless a lot nicer, but (and it's kinda hard to explain) nothing had been added to the game ... it seemed like it had been converted into 3D and re-released just for the sake of it. I'd have been just as happy playing it on my antiquated MD ... while the technology had progressed, the gameplay hadn't.
;) I don't hate FPS, but I generally only play them for a short amount of time, and I'm coming back to the 'why shell out for the latest and greatest console if what we've got will do' point.
;-)
The same strikes me with games such as Soul Calibre, the sequel (?) of Soul Blade on the PS. I know there's a very limited amount of things you can do with a Beat 'em Up, but it seemed to be another case of "we'll chuck a few new characters in and ponce up the graphics. *kerchiiing*".
What I'm trying to get across is the point that are games actually benefiting from more powerful consoles, other than in the graphics department ? Obviously, there's more that can be done with CPU time than polygons.
As for my comment about Goldeneye, I wasn't actually trying to suggest that it was superiour to Q/QII/Q3/Q3A/(insert name here), I was, albiet badly, trying to state that the graphics gave me no incentive to play the game, but that I thought it had good gameplay. (Add to this hours of fun playing in 'Big Head' mode, etc
As for the games in the wing for the X-Box, my knowledge of this area is poor (I'll take a look at the URL once I've finished this), so I'll take your word for it.
Your point about not having to program to a minimum spec is the one I most apreciated, and yeah, I can see this being a good thing. Just so long as MS don't start churning out X-Box SE or X-Box ME =)
Anyway, I'm off to dig out my BBC B for a game of DogFight
What, you mean like this ?
Personally, I'm afraid I can't see what's going to be so great about the X-box. I've read the specs, and wow, can it shift a lot of polygons ... but common sense suggests that chip makers other than nVida will have bought out darn powerful cards which I can put in my PC by next year.
... but that was pure playability, nothing to do with graphical wonders). Could I enjoy such greats as Baldurs Gate and Dialo 2 on an X-Box ?
While people are busy drooling over the number of polys this thing can put out, what about the quality of the games themselves ? Personally, I'm not keen on the Quake genre of games (well, maybe Goldeneye
It strikes me that the X-Box will just become another platform for First Person Shooters and Beat 'em Ups, and regardless of who it's manufactured by, this leads me to the conclusion : no thanks.
From the comments I've read, it seems that people are generally not happy with this move.
... after all, they're the creators of Solaris, so it's far easier and cheaper for them to support boxes running Solaris than it is for them to support Linux. It's like the MS/Hotmail thing. It was an embarasment that Hotmail ran on BSD, not Windows, and hence MS changed that. Likewise in this case; Sun have spent much mulah buying Cobalt, who's products use, what must be classed as, a competitor to Solaris.
.... first of all, if they do start shipping them with Solaris, is there anything stopping people replacing it with Linux themselves ?
Several people have already pointed out that the later RAQs are x86 (K6-2 if I remember correctly) based, and I agree with one of the above posts with regard to seeing Sun's point of view
Anyway, on to my point
And secondly, why so much anti-Solaris feeling ? I use both Solaris (at work) and Linux (at home). Yes, I see differences, but I don't see what's so bad about Solaris. So, instead of just slaggins Sun off, could people offer some constructive critisism about why this move would be a bad thing ?
Does this now mean that Ballmer has the power to give Mr Gates his notice ?
Do you think that Bill knows something Steve doesn't ? :) Also, seeing as Bill's been in the 'Opressive Marketing / Global Domination' department for so long, what does he hope to bring to company as 'Chief Software Architect' ?