I love it when outsiders to a community (like journalists) get things like this wrong, because it makes me rethink my mental categories.

We have absorbed a culturally agreed set of differences between a computer language, a platform, a framework, a library, an OS... and we're comfortable with licencing being coextensive with these units without getting confused, for example.

I'm not so sure about the library vs framework agreement. Many packages that would historically be labeled libaries, now call themselves a framework, making it even harder for the real frameworks to stand out among the clutter.

Isn't the difference generally that a library is used by client code, whereas client code sits 'inside' (for varying levels of inside) a framework? That's my rule-of-thumb anyway.

There's certainly a continuum, almost any decent library can be extended by user code.

That's how I used to remember things. You call a library, a framework calls you (is how I kept things straight when I started out in CS)

This phrase, spoken by someone who doesn't know anything about the subject, contains an unrelated grain of truth.

Today's projects are cobbled together by giants sections of open source code, customized for specific purposes. A project may span languages, and usually includes more than one.

His ignorance made me appreciate the meta-programming language one could call "open source code."

To be fair when I talk about, say, legal issues I tend to blurt out dangerous half-truths that I deduct from supposed 'first principles'. On the other hand I would never write an article on the law like this person has done.

This makes me wonder whether all the articles out there on other stuff contains errors like this too.

Michael Crichton gave a name to the absence of such wondering:

Briefly stated, the Gell-Mann Amnesia effect works as follows. You open the newspaper to an article on some subject you know well. In Murray's case, physics. In mine, show business. You read the article and see the journalist has absolutely no understanding of either the facts or the issues. Often, the article is so wrong it actually presents the story backward-reversing cause and effect. I call these the "wet streets cause rain" stories. Paper's full of them.

In any case, you read with exasperation or amusement the multiple errors in a story-and then turn the page to national or international affairs, and read with renewed interest as if the rest of the newspaper was somehow more accurate about far-off Palestine than it was about the story you just read. You turn the page, and forget what you know.

http://web.archive.org/web/20070629011206/www.michaelcrichto...

They do. If you've ever had anything of yours reported in the press, you will discover that almost all articles are like this one.

It doesn't have to be that way

It's all the editor. If the editor accepts stuff like this, then it happens. Sometimes you luck out and get a writer who really understands, usually you do not.

Well, I'm sure it doesn't, but it mostly is. Suggestions?

ArsTechnica seems to be getting it right, and the other day I heard NPR call up one of their editors as an expert, so the mainstream news is noticing something.

I think the real problem is, most consumers don't know what they're missing.

Reinvent journalism from the ground up?

Many journalist have invented their own language known as "the newspaper" to be sold public streets at low cost, but profitability issues are murky.

Why would this guy bother copying stuff off the Goldman servers when it was open source, and presumably freely available?

I think that he claimed that the code was open source, but in fact it was proprietary.

Reading between the lines, this is my guess as to what's going on:

Goldman Sachs built an internal-use-only tool that relied on some GPL code. Aleynikov is going to claim that this made Goldman Sach's code fall under the GPL, and therefore he has a right to redistribute it.

I hope that's not what's going on, because the pointy-haired ones don't need another reason to be afraid of FOSS.

That is not how GPL works. Let's say you have derived work from GPLed source (DW).

if (you publish binary of DW) then {you have to provide sources of DW under GPL}

Simply it does not apply to in-house (non-published) work. You are fine to keep it closed-source.

I believe you are right.

This article is more coherent.

http://www.fiercecomplianceit.com/story/aleynikov-trial-and-...

It seems like he's essentially grasping at straws after the fact to defend himself. But as has been said, that's not how open source works. I don't think this is actually going to make anyone nervous since clearly no one bought his story. You can lie and not be believed about proprietary licenses too.

... I feel sad actually sympathizes with the prosecution here but that's how it goes.

Or maybe Goldman Sachs built a bunch of stuff internally on top of open-source code.

In which case, they could have been in violation if it was GPL code.

Generally, you're only required to provide changes if you are distributing your modified program. This may have been partially addressed in GPLv3 in the form of an "application service provider" clause, but I don't recall

I doubt they even fell under the "application service provider" clause; speed-trading software is usually used completely internally by a financial firm to make its own trades.

IANAL, but I think that's the AGPL, not GPLv3.

Not necessarily, as GPL3 didn't close the so-called "ASP loophole". Since Goldman Sachs probably wasn't distributing any HFT applications, but just selling access to it or using it completely internally, they aren't required to make their modifications public. See http://stackoverflow.com/questions/2127246/difference-betwee... about the AGPL license which addresses the 'application service provider' issue.

Really, what part?

GPL doesn't force you you to publish, just to distribute source with all binaries.

A bank with the proud history of Goldman Sacks cheating like that - unthinkable!

I think it was mostly oss (maybe with local changes, re: why not download a fresh copy?), but there was some proprietary code in there (which may or may not have been intentional), which is what tripped him up.

Right, but aren't we sort of underestimating GS's ability to tag oss with just about anything proprietary-ish, then, when one of your top programmers quits to join (or co-found) a firm in a competing industry (H-FT), maybe hold it over their shoulders. Considering the scope of this and the dollars to be made, Mr Aleynikov should have been aware of his former firm's contractual small print and should not have copied anything without an IP lawyer present.