Star Microsoft menaced by Open Source
Linked by David Adams on Tue 5th Aug 2008 21:20 UTC, submitted by JCooper
Microsoft Microsoft . . . complained in its annual report that it was facing increasing pressure from open source companies. It claims they are stealing its ideas and benefiting from its intellectual property. "A number of commercial firms compete with us using an open source business model by modifying and then distributing open source software to end users at nominal cost and earning revenue on complementary services and products." Also see analysis at Microsoft Watch.
E-mail Print r 0   58 Comment(s) http://osne.ws/fjt
Thread beginning with comment 325811
To view parent comment, click here.
To read all comments associated with this story, please click here.
lemur2
Member since:
2007-02-17

The trends in commercial software are moving more and more to user driven development, and not compromising on quality. Where I work we have a 1 month release cycle, where every month we ask our users (we call them product owners internally to try and keep from falling into holier then thou programmer attitudes) what the features they need the most are, and let that drive the features of the release. We have two testers to every developer, and these are trained people who give very good bug reports and methodically test the software, not just random people on the interweb. We aim for 3 lines of test code to every 1 line of production code, and most of the team practices test driven development. Every code check in needs to be reviewed by another programmer before it goes to the testers, and our build server automatically runs through integration tests on every check in to the source tree. And every iteration has time set aside for regression testing, deployment testing, and performance testing to be absolutely sure we ship quality.


All of this is required if you are about to sell a product in exchange for money.

Open source code isn't about that at all ... it is about exchanging code in an effort to evolve a better product.

If I am but a lowly "random person on the inetrweb", and I download a piece of open source software (say KDE 4.0, for example), and I read and understand that I am being offered the software at no cost to me but with no guarantees as to its performance, and I then try out the software, notice a problem with it and I submit a bug report ... have I not tested the software?

Even though I am just a lowly "random person" the bug I discovered is still a bug no matter who discovered it. It does not become a bug just because software QA noticed it.

Your prejudice is showing here.

Reply Parent Bookmark Score: 4

google_ninja Member since:
2006-02-05

If I am but a lowly "random person on the inetrweb", and I download a piece of open source software (say KDE 4.0, for example), and I read and understand that I am being offered the software at no cost to me but with no guarantees as to its performance, and I then try out the software, notice a problem with it and I submit a bug report ... have I not tested the software?


No, you have discovered a bug. Testing the software is boring as hell, because it is basically hammering at the interface in a methodical fashion to hit every combination of inputs possible. It is also very effective, because given enough users, chances are those combinations will get hit eventually.

You could argue that with enough users, software will be well tested, but that would be considered absolute lunacy in every single other industry in the world right now, except for some reason, programming. The whole point is to address these things before the lemur2's of the world get their hands on it.

The only people who even come close to being testers in the open source world are those who run trunk builds on a daily basis, are active on the mailing lists with feedback, and submit proper bug reports. Everyone else is just users of untested software

Reply Parent Bookmark Score: 5

lemur2 Member since:
2007-02-17

"If I am but a lowly "random person on the inetrweb", and I download a piece of open source software (say KDE 4.0, for example), and I read and understand that I am being offered the software at no cost to me but with no guarantees as to its performance, and I then try out the software, notice a problem with it and I submit a bug report ... have I not tested the software?
No, you have discovered a bug. Testing the software is boring as hell, because it is basically hammering at the interface in a methodical fashion to hit every combination of inputs possible. It is also very effective, because given enough users, chances are those combinations will get hit eventually. You could argue that with enough users, software will be well tested, but that would be considered absolute lunacy in every single other industry in the world right now, except for some reason, programming. The whole point is to address these things before the lemur2's of the world get their hands on it. The only people who even come close to being testers in the open source world are those who run trunk builds on a daily basis, are active on the mailing lists with feedback, and submit proper bug reports. Everyone else is just users of untested software "

Again you fail to understand, so again I will patiently try to explain it to you.

Open source code release is not delivery of finished product to consumers in exachnage for money.

Open source code release is collaborative field testing of product in development.

Since the code is tested far and wide by millions of participants in a wide variety of field conditions, by the time it has become mature (called a "stable release") it is far better tested, and greater quality, than closed source code tested only in-house by the same company that produces the code in the first place and which is (as a whole) under considerable market pressure to release product.

Case in point ... Vista.

QED.

Reply Parent Bookmark Score: 6

Read more of this thread... >
_txf_ Member since:
2008-03-17


The only people who even come close to being testers in the open source world are those who run trunk builds on a daily basis, are active on the mailing lists with feedback, and submit proper bug reports. Everyone else is just users of untested software


The thing is that most open source devs eat their own dogfood, so will be among the people who test and find bugs.

let me correct you:

"Everyone else is using tested software"

The quality of that test is up for debate, but you cannot say that there are testers and then say that the software is untested.

Reply Parent Bookmark Score: 4