On Tue, Sep 29, 2009 at 8:00 AM, Jason Grout <jason-s...@creativetrax.com> wrote: > > William Stein wrote: >> Hi, >> >> We (mainly me and Tim Dumol) made a lot of progress getting the >> standalone notebook to work, and we're going to switch to the >> standalone Sage notebook in sage-4.1.2. > > > This makes me very, very nervous.
Good. > Why the rush? There's not a rush. > I thought your other > idea of keeping the current notebook for 4.1.2, but including this > notebook, and asking lots of people to try this out, was a much better > and safer alternative. The notebook as it is in 4.1.2.alpha has a bunch of bugs that were introduced in the patches for 4.1.2, with many things applied on top of them. We fixed all of those bugs in the separated notebook. Moreover, the security model of the separated notebook, for people using server_pool, is dramatically better. Right now, any user can trivially delete everybody else's work on sagenb.org, say. With the new notebook this is not be possible -- the "worksheet process infrastructure" is much more well designed. So I think the separated notebook will be a major improvement. > How about making it so that notebook(new=True) gives the new notebook, > but otherwise gives the old notebook, just for 4.1.2? That will give > many more people a chance to try it out and report bugs. > > For example, here's another bug: > > Copy your "a random test worksheet full of stuff". > > Go to "Data|foo.txt" > > Change the text > > Click "Save Changes" > > An error page appears. Thanks. We'll fix that. > It's these corner cases that I think are just going to take time and > lots of people to track down. There is time and there are a lot of people reading this and testing out http://uw.sagemath.org. This is a list of everything those people have been able to find wrong with uw.sagemath.org to date: [ ] reimplement sphinx introspection. * confdir in cell.py won't work without a trivial Sphinx directory, when not in Sage. make it. [ ] "Sign into the Sage Notebook v" <-- missing version number [ ] Copy your "a random test worksheet full of stuff". Go to "Data|foo.txt"; Change the text; Click "Save Changes" An error page appears. Anything else? > Pushing this in at this point I think is > premature, given how much people rely on the notebook as their single > gateway to Sage. It's not being pushed out today. And anyways keep in mind that this isn't a total rewrite of the notebook. It's just the exact same notebook code copied out of sage into another location, with some additional work done to make this work independently of Sage and not depend on the pexpect interface code. > When is the release date for 4.1.2? There is no date yet. The two main blockers are: * OS X 10.6 support * the separated notebook If the OS X 10.6 port would have been finished last week, I would be OK with releasing Sage asap. However it is not done -- there is a major blocker problem there involving pyCXX being completely broken on 10.6 (I fixed all the other problems though last week). As far as I know, nobody has any clue how to fix this problem and nobody is even working on it. I've emailed various people. If nothing happens, I'll try myself later this week. William --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
