gmane.org.fsf.europe.discussion

Re: Stallman: cloud computing is a trap

Adam Sampson — 2008-10-06T21:51:44

I suspect there are lots of informal arrrangements along similar lines

Re: Stallman: cloud computing is a trap

MJ Ray — 2008-10-06T16:54:08

Pragmatically, what is the difference? It may be under a free software licence, but "Access to the source code is a precondition" for "The freedom to study how the program works, and adapt it to your needs" and "The freedom to improve the program, and release your improvements to the public, so that the whole community benefits". http://www.fsf.org/licensing/essays/free-sw.html I don't care much if something is theoretically free software under copyright law if it's not practically free software. Back tomorrow,

Re: Stallman: cloud computing is a trap

MJ Ray — 2008-10-06T16:50:23

I don't understand this question. I was asked what proprietary webapps those FSF projects were using and tried to offer a summary. ODP aka dmoz.org and not entirely, but it was the best I found. I also sell some of my searches to a proprietary engine in exchange for meal vouchers, cinema tickets and so on, but I don't promote it. I don't value my search data much - my bookmarks are my first stop. Where those are linked from www.gnu.org as official parts of the development effort, that would be similar to how it seeks to enforce the naming, conduct and activities of all user groups linked. Personally, I'd prefer it to standardise on liberalisation, linking freely and not advocating Affero. What I'm really pointing out here is that FSF's stance on third-party resources is unpredictable. Ow, that page is dated and the software is obsolete - parts survive in the bizarrely-named schycyrssmerge2, which is a fairly efficient aggregation engine based on the set theory, but it's pretty niche. Thanks for t

Re: Stallman: cloud computing is a trap

Alfred M. Szmidt — 2008-10-06T16:37:14

Just because _you_ don't have the soruce code to something, does not mean that something is non-free software. While there are issues of privacy, it does not mean that for examople FSD is running on non-free software.

Re: Stallman: cloud computing is a trap

Noah Slater — 2008-10-06T16:00:47

If you could avoid rhetoric and non-value neutral phrasing it would go a long way to promoting intelligent discussion. By imposing your value system through your language you attempt to influence the reader without citation or argument. [...] So, by talking about, linking to or externally active with a third party website you consider this to be a violation of *your* principals? Can we not say "free network software is great, let's do more of that" instead of trying to imply that it is even possible to use the Web as it stands without using some "non-free" network service. Tell me, which search engine do you use and is it free software? My apologies, this was a misreading on my part. Not really. The GNU Project is a volunteer effort, are you asking that it enforces the conduct and activities of all it's maintainers external to the official infrastructure? I find this particularly hard to swallow from a fellow Debian contributor who presumably toes the party line that Debian's contrib and non-free

Re: Stallman: cloud computing is a trap

MJ Ray — 2008-10-06T15:35:37

Even if the lawyerbombs resolve in the best possible ways, it pushes up costs of these marginal providers because of the sins of Big-Webmail-like providers. Yes, as discussed previously. Is digg free software? I forget. I think this is one of the better FSF projects. Uses facebook. I don't like the general thrust of that project either. Depends on one's opinion of AGPLv3. Data is verbatim-copying-only. Wikidot doesn't offer users an export option, as far as I've seen, so is it free and open in practical terms? Not seen those and no time to check now, sorry. I don't remember what FSD is running and it's not on the site. It doesn't seem to be open source, let alone free in a practical way. The data is under the FDL, which isn't a free software licence. I know many on this list think that doesn't matter, though, so please don't remind me of it ;-) It does make it rather impractical to use the data on other sites if a copy of the whole licence has to be in each derivative work... Noah Slat

Re: Stallman: cloud computing is a trap

Hannes Hauswedell — 2008-10-06T15:28:12

We have had a similar discussion on fsfe-de and I dont want to repeat all the points, but I think we really need to differentiate between to issues here: 1) Where do I store my data, do I trust the people in charge? 2) What freedom does the Software involved offer me? The only point where these issues mingle, is when the software is not controlled by the same people controlling the data (e.g. the software is proprietary and provided by a third party), resulting in another party of people you would have to trust. Both of these issues can be adressed individually or together, but they are two issues, e.g. you can de-/encrypt data on the client-side and store it an proprietary web-app solving issue 1, but leaving issue 2 open. Or you can save unencrypted data in a AGPLed CMS on GNewsense-Server and still have the people running the server sell your data. Now, what we have to think about, is whether both issues are equally important to soceity in general and how or if the Free Software Movement shou

Re: Stallman: cloud computing is a trap

Matt Lee — 2008-10-06T14:02:41

Re: Stallman: cloud computing is a trap

Noah Slater — 2008-10-06T12:53:53

Please clarify this statement. The GNU website is available via CVS. These are used by the individual software developers working for GNU and are not hosted or part of the official GNU website. Who?

Re: Stallman: cloud computing is a trap

MJ Ray — 2008-10-06T12:06:00

I was thinking mainly of the off gnu.org homepages of some GNU software. I think that maybe sourceforge is justifiable, particularly where its use predates the existance of Savannah and other free/open hosting sites, so I wasn't thinking of that, but there are other parts of the GNU project using things like Atlassian Confluence - it's not like we're short of free software WikiEngines. So it wasn't what I was referring to above, but I do feel that AGPL is probably non-free, but it depends how the vague bits are resolved. However, I fear that all AGPL-using sites would be blanket-OK'd by FSF, which is why I think they wouldn't be a good rating agent for webapps. Also, some of the strongest AGPL advocates are heavy users of proprietary web-apps, which is particularly irritating, like a drunk lecturing on how we should all be sober at all times. Hope that explains,

Re: Stallman: cloud computing is a trap

Ciaran O'Riordan — 2008-10-06T11:40:20

MJ Ray public.gmane.org> writes: I don't know which proprietary webapps you're refering to. Do you mean "proprietary" as in FSF violating it's own interpretation of free software or violating your definition (you might call AGPL non-free) of it? Or do you mean that not FSF but the off gnu.org homepages of some GNU packages are using proprietary webapps (like maybe sourceforge?) ?

Re: Stallman: cloud computing is a trap

MJ Ray — 2008-10-06T09:21:43

Welcome. I include my answers to some questions in-line below. Could you correct your flow/line-length, please? I got very long lines here, which is a pain to reformat before replying. Another type of provider is a web-application user cooperative: groups of users gathering together to employ people who can maintain a webserver and application(s) for them. There are currently very few of these and all the ones I know about serve limited communities. It's not difficult to start one and local Cooperative Development Agents will probably advise. Follow links of http://www.ica.coop/ As I understand it, by EU law, employers have to allow reasonable private use of personal (rather than role-based) email addresses, just like they have to allow telephone calls. They can lock it down, but it must not be "without my knowledge". One of the big problems with Big Webmail is that they are often in a foreign country, governed by sometimes very weak privacy laws, and sometimes with even weaker privacy terms that

Re: Understanding MPL

2008-10-06T08:42:21

Thank you very much, that was really helpful. Now, from what you said, also this issue seems pretty clear. A derived class, that is inheriting attributes and behaviour from another class is a derivative work. However, I think, since the derived class normally is in a separate file only the LGPL requires it to stay LGPL. The MPL should allow another license for that file/class, even though it is strongly connected to - and definitely does not make sense in any way without - its super/father/base class. Regards, gk

Re: Stallman: cloud computing is a trap

Noah Slater — 2008-10-05T22:02:43

Hello, I think this is very apropos: http://autonomo.us/2008/09/rms-on-cloud-computing-stupidity/ I urge you all to read up about autonomo.us and the Franklin Street Statement. http://autonomo.us/2008/07/franklin-street-statement/ Best,

Re: Stallman: cloud computing is a trap

Ciaran O'Riordan — 2008-10-05T21:19:02

Here's an article about how to include freedom in web applications, including input from Stallman: http://www.clipperz.com/users/marco/blog/2008/05/30/freedom_and_privacy_cloud_call_action I think Wikipedia is a good example of a good, freedom respecting web application, but it's a rare exception. A web app could be ok for email if: 1. Email was stored encrypted in a way that only you could read it 2. You could download all your data (including meta-data such as settings, aliases, groups) and move to another computer without too much difficulty 3. You had control over your email address. You can do this by buying your own domain or using an alias provided by an organisation you trust (like the fsfe.org alias I use)

Re: Stallman: cloud computing is a trap

Carsten Agger — 2008-10-05T18:08:50

søn, 05 10 2008 kl. 13:38 +0200, skrev Thomas Pfeffer: I'm just a private citizen and can't speak for anybody but myself, but here's a few ideas: Technically, community web applications and web services (not Web-2.0 behemoths like Google or Flickr) - might be built on something akin to TOR or bittorent - everybody participating might donate some bandwidth and (say) a gigabyte of space, and the data might be encrypted and accessed through decentralized, anonymous peer-to-peer processes. * You own your data and they may never be accessed for other purposes than designated by you (violated by Google, who skim your emails to serve ads - violated by Facebook and others, in that they presume "ownership" over everything you write and all pictures you post). * You must be able to encrypt your data and decrypt it on the client side, i.e. without ever supplying the private key to the service provider. (Violated by all Web 2.0 applications) * You must be able to delete your data, and when you've deleted them,

Stallman: cloud computing is a trap

Thomas Pfeffer — 2008-10-05T11:38:45

Re: Understanding MPL

Ciaran O'Riordan — 2008-10-01T23:11:45

The MPL is often called a file-level copyleft licence. I haven't read the MPL in years, but a summary I heard from a reasonable person is that if you make changes in the files you received, you have to keep the same licence that whole file. But if you make some changes to existing files, and add your own new files, then you can freely choose the licence of your own new files. "gk-VaTbYqLCNhc< at >public.gmane.org" public.gmane.org> writes: By what I wrote above, yes. By what I wrote above, yes. The MPL is much weaker than the LGPL. All changes to LGPL'd software has to be released under the LGPL - even if the changes are in a separate file, if they form part of the thing covered by he LGPL, then they have to be LGPL'd. Only things that could reasonably work without any of the LGPL'd code can be considered separate and not bound by the LGPL. The LGPL also has good safety features to ensure that when someone receives a combination of LGPL'd software and proprietary software, the user can

Understanding MPL

2008-10-01T14:35:12

Hi! I am currently reading the MPL 1.1, and have some basic problems understanding it. --8<--------------------------------------- 3.7. Larger Works. You may create a Larger Work by combining Covered Code with other code not governed by the terms of this License and distribute the Larger Work as a single product. In such a case, You must make sure the requirements of this License are fulfilled for the Covered Code. --8<--------------------------------------- What does "combine" mean? Especially when modifications are defined as --8<--------------------------------------- 1.9. "Modifications" means any addition to or deletion from the substance or structure of either the Original Code or any previous Modifications. --8<--------------------------------------- (1) Does that mean, that as long as somebody does not use lines from a file of the Covered Code, he may distribute his proprietary code, which calls functions of the Covered Code, in a single executable? (2) Only when he alters the Covered Code, or

1st Fellowship interview: Seán Daly

Ciaran O'Riordan — 2008-09-26T12:55:11

We've just started a series of monthly interviews with the Fellows: http://fellowship.fsfe.org/interviews Seán Daly is the first interviewee: http://www.fsfe.org/en/fellows/ciaran/fellowship_interviews/fellowship_interview_with_sean_daly (If passing this on, please also spread the first URL) Seán talks about how he got into doing interviews as a volunteer, and why he was particularly interested in the EU vs MS anti-trust case. We also cover DRM and other topics. You can vote for it on these sites: http://digg.com/software/FSFE_Fellowship_interview_with_Sean_Daly http://www.fsdaily.com/Community/FSFE_Fellowship_interview_with_Sean_Daly

Re: Printable information material available for download

Reinhard Mueller — 2008-08-31T23:13:58