On Fri, Aug 15, 2014 at 6:02 PM, Thierry <sage-googlesu...@lma.metelu.net> wrote: > Hi, > > On Fri, Aug 15, 2014 at 08:52:03AM -0700, Harald Schilly wrote: >> >> This should be moved to sage-flame. >> >> On Friday, August 15, 2014 10:42:14 AM UTC+2, Dr. David Kirkby (Kirkby >> Microwave Ltd) wrote: >> > >> > If not against the GPL, this certainly seems to be going against the >> > *spirit* of the Sage project. >> > >> >> What's the spirit of the Sage project? It's mission statement? > > The mission statement is about a "free open source alternative", which SMC > is not.
Yes. However, SageMathCloud != Sage, in fact, far from it. The closest analogy I can come up with is github vs. git. Github is a for-profit web service run on closed-source software that owes its existence to the completely open source git. On the balance I would say that its existence is a net positive for the git project. The (lack of) lock in story is much easier and more transparent with github, but I am hopeful that SageMathCloud will become similar. >> Please explain. In particular, how making an almost unmodified version >> of Sage available to a much larger user-base with a greatly reduced >> level of entrance (i.e. creating a login credentials vs. downloading >> >1GB, installing VirtualBox and running a virtual machine) has any >> >implications besides fulfilling the mission... > > The user-base could be even larger if every university (or whatever) could > host its own local cloud and freely adapt it (e.g. translations, ldap > interface, ...). We had this in the form of the original sage notebook for years. Yes, it had its problems, but "universities" didn't step up to fix them (or replace it with something better). Few others than William hosted such servers of significance. I may have drunk the cloud kool-aid, but I don't think most people actually want to manage their own local clouds (though I'll admit that wanting to be *able* to, despite not wanting to actually do it, is a legitimate request). Dealing with the whole stack--everything from hardware up to spammers--is a lot of work and it often makes sense to hire this out. >> > Maybe it is legal. I don't think it is morally right. >> > >> >> Everyone contributing to Sage is hopefully aware that it is GPL >> licensed. The core essence of GPLed code is that it can be used for any >> purposes - free (as in beer), commercial, medical, military, energy, >> etc. [0] For me it would be morally questionable, if e.g. you would >> impose moral considerations about how the code should be used. In >> particular, who gives *you* the legitimacy to decide what is a moral use >> and what is not a moral use? Nobody should have that. > > By the closed-source nature of SMC, the University of Washington can > impose moral considerations to the users of SMC : if UW considers that > some user does not use SMC in a moral way, it can delete its account, and > the user will not be able to connect elsewhere to still enjoy its > features. This is not a purely closed vs. open source issue--letting someone else host the service even if the entire stack were open exposes you to this danger. (If it were open sourced, you could theoretically take your pick of providers, or run a cloud yourself, but given the experience with the original sage notebook I don't know that this would actually happen very often.) While I don't find the "you can't use our service any more" inconvenient, the only thing I have a visceral reaction to is loosing access to all your data. I'd like SMC to have a clear story around this similar to Google's Data Libration Front. The "personal use" SMC plays into this as well. That being said, I too would like to see more of SMC released under an open source license, and sooner. But to me the data story is even more urgent. The fact that project histories are stored in a git-like format means that this shouldn't be too hard technically. >> [0] freedom 0, see https://www.gnu.org/philosophy/free-sw.html >> >> My own point of view is that there are great examples where using >> open-sourced code for different circumstances improves it's quality. >> e.g. Linux Kernel, Mozilla Project, … My core argument is, that merely >> implanting Sage in different environments doesn't diminish its value or >> functionality. > > The environments could be even more diverse if SMC was free software and > widely spread, its quality could be improved if it could be run on various > clusters. > > Your arguments are good, but they all support the open-sourcing of SMC, > which was Dave's point (if i understand correctly). > >> Finally, if you do not like SMC, ignore it. > > As long as there is a huge advertising of SMC on the sagemath.org front > page, and since the development of SMC is (partially) funded by the Sage > foundation (whose aim is "To support the development of the mathematical > software system SAGE."), this is not really possible. There is also the huge opportunity cost that William Stein, and those he's hired, have spent the last several years developing SMC rather than Sage itself. The jury is still out on whether that's a net win, but I'm cautiously optimistic that it will be. Coming back to the original mission and spirit of Sage, open-source makes using this software possible, but doesn't make it easy which is an essential part of the equation we need to solve. A hosted service goes a long way to satisfy help the easy-of-use part of the equation, and I'm skeptical that this can be done at scale through grants, volunteers, and benevolent university IT departments. And should SMC become profitable from a financial point of view, I would hope (expect) it to be structured as a non-profit funneling proceeds to fund the open source math software development side of things rather than private yachts, and the prices geared around providing access in self-sufficient manner rather than maximizing profits. - Robert -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
