slabbe wrote: > Hi, > >> To me, a year is FAR too short. Mathematica makes a major release about once >> every two years, and a semi-major one every 6-12 months. There are plenty of >> people using older versions of that. If Sage wants to make a viable >> alternative >> to Mathematica, it needs to keep deprecated functions a lot longer than a >> year. >> I know WRI do - I'm also aware they break backwards compatibility some times. >> >> I think 10 years would be best. Any less than 5 is asking for trouble. Sure >> Sage >> makes a new release every couple of weeks, but as we see from recent >> discussions, not everyone wants to update every couple of weeks. > > Wow, 10 years seems, to me, far too long. Do that means that we think > that somebody is currently writing code in Sage 4.2.1 and will then > update to Sage 15.0 in 2020 but his code will be broken because > deprecated functions will just have been removed? Is there another > example for suggesting 10 years?
If Wikipedia is to be believed: http://en.wikipedia.org/wiki/Solaris_%28operating_system%29 then Sun are supporting SunOs 5.8 (Solaris 8) until 2012, which is 12-years after it was released. Solaris 9 is supported until 2014, which is also 12 years since it was released. I would add Solaris is also free. I'm also the first to agree that updating an operating system has more risks than updating Sage, but on Solaris at least, it is less time-consuming! > It would be interesting to get statistics for frequencies of updates > for all the Sage users. I would think that the frequency is higher for > Sage than it is for other commercial software simply because it is > free to do it. If I paid a lot to get Matematica X, I might wait more > n years to buy an improved Matematica X+n. Well really, if you pay for Mathematica, and intend having any updates at all, it is cheaper to pay maintenance every year. > LTS Ubuntu are supported 3 years for desktop and 5 years for server > whereas usual version of Ubuntu are supported 18 months. See notes above some other software used in a professional environment. I know some of you linux guys probably find Solaris a pain, but it looks like Oracle is going to be buying Sun for 7 billion USD. For a company to be valued at $7,000,000,000 they must have done something right! > Anyhow, I would suggest that deprecated function be supported for at > least 2 * the average period of update made by Sage users. Can > somebody estimate this period?!! I'm not a statistician, but would breaking code for only 1% of users not be somewhat more sensible? i.e 5x the standard deviation? Twice the mean upgrade rate would to me seem far too short. (In practice, I doubt it is a 'normal distribution' so perhaps this is not valid. Anyway, you probably get my point.) I personally think a commitment to try whenever possible to not break backward compatibility for 10 years would certainly go some way to giving a professional impression, and so increasing usage in a professional environment. Since Sage is made up of numerous components, we do have far less control over these things than Sun or Wolfram Research. Personally I'd like to see Sage used more in non-academic circles more. Updating Sage and finding something you what done only 3 years ago does not work, would I feel be very annoying to such users. It would also mean something one starts at the beginning of a PhD could stop working before you finish, if the system admin updates Sage. That I personally find unacceptable. This is going to be one of those things where there is no 'right answer'! Dave --~--~---------~--~----~------------~-------~--~----~ To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
