Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).

×

Comment: Re:Well, this just screwed the legal pooch... (Score 1) 225

by Paul Jakma (#49432503) Attached to: How Ubiquiti Networks Is Creatively Violating the GPL

The GPLv2 is not perfectly happy with DRM. It is very much possible to read the GPLv2 as requiring *all* material needed to install an executable, which would include keys:

“For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable.”

The GPLv2 was very much intended to allow end-users to be able to *install* modified works. The incident which motivated RMS to start this whole free software thing and come up with the GPL was a printer whose software he wanted to fix but couldn't. The freedom to modify software on hardware you own is what the GPL was intended to provide.

The GPLv2 is *not* "happy" with DRM. At best, this is an untested grey area simply because the GPLv2 predates the notion of DRM and so could not use the language we use today. However, it clearly intended to cover installation. The GPLv3 unambiguously fixes this wording issue. That does not mean the GPLv2 allows it though.

Comment: Re:There's a limit to what can be done (Score 1) 385

by Paul Jakma (#49356613) Attached to: Modern Cockpits: Harder To Invade But Easier To Lock Up

How many suicidal hijackings have there been? 3 - and the passengers had already realised on the day and changed their behaviour as #3 was in progress. The best defence against these hijacks was already in place by September 12th - the passengers.

How many suicidal pilot crashes have there been of jet airliners? At least 4.This one, Egypt Air 990, Mozambique Airlines TM470, Silk Air Flight 185, since 1999.

There's just no good defence against a suicidal pilot. Bear in mind that Egypt Air 990 went from level flight at FLA330 to the ocean in 43s, even thought the captain was back in the cockpit within 12s (no locked door then) and behind the controls no later than 27s.

Also bear in mind cockpits have at least one crash axe (for crashes and getting at electrical fires behind panels).

IMO the best thing to do would be to get rid of the locked doors. Whatever chance the other pilot has, it would surely be increased if other crew and/or passengers could easily and quickly get in to assist with any man-handling needed.

Comment: Re:Ummmm ... duh? (Score 3, Informative) 385

by Paul Jakma (#49356389) Attached to: Modern Cockpits: Harder To Invade But Easier To Lock Up

Gender is irrelevant. Egypt Air 990 crashed *without* locked cockpit doors. The captain was back in the cockpit within 12s of the co-pilot initiating a descent. He was making control inputs within 27s. However, he didn't start to suspect the cause of the problem might be the co-pilot until between 30s to 33s. The aircraft hit the sea at about 43s.

Every second may be vitally precious in these situations. Locked cockpit doors, even with over-rides, will waste potentially extremely-critical time.

Comment: Re:I'm dying of curiousity (Score 1) 188

by Paul Jakma (#49216975) Attached to: Software Freedom Conservancy Funds GPL Suit Against VMWare

Except they didn't re-implement the sub-system framework and data-structure APIs themselves. The lawsuit alleges that they took some code from Linux (e.g. radix tree, timer API stuff). Even if VMWare /had/ reimplemented those APIs from scratch, then there is still the issue that they /also/ have appropriated the code GPL-only drivers (as alleged by the lawsuit at least) for ESXi.

Also, will you provide that list of drivers? If you won't, I have to wonder if you're favourably predisposed to VMWare in some way.

Comment: Re:I'm dying of curiousity (Score 1) 188

by Paul Jakma (#49194837) Attached to: Software Freedom Conservancy Funds GPL Suit Against VMWare

BTW, a vendor that wrote a Linux driver could give a different, non-GPL licence to that driver code, of course. However, that still leaves the issue that Linux drivers are written within a framework of core Linux code (driver sub-system specific frameworks and further more generic services and data-structures). The driver vendor can not give a non-GPL licence for that core code they didn't write.

VMWare are alleged to have copied such core code too. Further, they are alleged to also have used GPL driver code (e.g. Hellwig's SCSI). So VMWare, according to the allegations, have borrowed GPL code on /both/ sides of the "line" between drivers and their supporting code. Without fulfilling the conditions required by the GPL for legal, licensed use of GPL code...

Comment: Re:I'm dying of curiousity (Score 2) 188

by Paul Jakma (#49194815) Attached to: Software Freedom Conservancy Funds GPL Suit Against VMWare

The Software Conservancy FAQ has a diagramme giving an abstract of what they allege has been copied:

http://sfconservancy.org/linux...

With respect to the drivers, it seems they've copied SCSI, USB and network drivers. Christoph Hellwig holding copyright to at least some of the SCSI drivers concerned, in addition to core code VMWare are alleged to have copied to implement required APIs for the drivers.

It seems you could give us a full list from an ESXi installation, if you wished, rather than just a selected driver (selected why?).

Comment: Re:I'm dying of curiousity (Score 4, Insightful) 188

by Paul Jakma (#49194509) Attached to: Software Freedom Conservancy Funds GPL Suit Against VMWare

My understanding is that your description is inaccurate.

Yes, they have implemented a number of Linux APIs in their own code. Additionally, they have sucked in bits of GPL Linux code that implemented bits of those APIs (i.e. NOT reimplemented, as WINE does). This is to allow ESXi to be able to re-use drivers from Linux, as you say. However, they didn't stop there, from what I understand. They also have ESXi use Linux GPL drivers.

My understanding, from what I've read, is that ESXi didn't just re-implement Linux APIs. ESXi also heavily depends on GPL licensed Linux code, both in the partial-reimplementation of Linux APIs, and in sucking in Linux GPL drivers. The issue is this direct re-use of GPL code, and ESXi's heavy dependence on that GPL code. That dependence likely makes ESXi a derived work of the Linux GPL code, and as such it - in its *entirety* - must be distributed in accordance with the GPL.

Alternatively, VMWare are quite free to not use code they didn't write and don't own, if they don't like the licence conditions.

This is very different from what WINE does, and to characterise the situation as like WINE seems to be quite inaccurate.

Theory is gray, but the golden tree of life is green. -- Goethe

Working...