Thanks Rob. It is great to see I am not the only person who has gotten excited. This idea is soooo good I am having trouble finding a reason to not Just Do It TM.
I know there is a Sage Journal idea floated (JSage). So some of what I am thinking is surely motivated by that. Here is what I envision: * A front end like arxiv.org, but with a distinguished board of editors who organise reviewers for the papers, i.e. a proper journal setup, but with an algorithmic/computational bent. * The archive would provide a list of papers, which when selected would send the worksheet to one of a list of sage servers set up around the world. Each server might have different Sage/GP/Magma, etc binaries installed on it. The paper would specify which binaries it relies on and the archive mechanism would only list servers which provide those binaries. So you'd effectively select a paper and a server and the process of bringing that paper up for you would be automated. (This would mainly be for papers including benchmarks - for other papers a minimum or list of binaries supporting the examples in that paper would be provided). * Universities would pay for a subscription to the service, which would pay for the hardware to run all the different CAS binaries which would be required, and pay for work on the notebook itself, maintenance of the archive system, storage, etc. A University might get a free subscription if they provide a server. Individuals might be able to pay for a lifetime subscription just for themselves, for a nominal fee, which they could pay for out of grants, etc. Non- academics not working for companies who could pay for their subscription could write to the maintainers for a free account. I would hate to absolutely limit access to ideas/papers, but the infrastructure required to set this up would need to be paid for somehow, and since the whole thing has come from the Sage process, I don't see why the Sage Foundation could not provide jobs for people to do work on Sage and the notebook as a consequence of such funding. * There could also be a selection of refereed expositions available, which would be required to have examples implemented in some supported CAS within the worksheet. * Papers could contain live benchmarks, and getting up-to-date benchmarks would be as simple as running on a server with more recent binaries (this would be an available option at the archive level). The only problem I can see is that when parallel processing becomes the norm, very large computations and benchmarks might be included in papers. Obviously the execution of those would have to be limited. Perhaps an option could be provided for a researcher to re-run such computations on their own Sage server, so they are then responsible for providing the cycles. I guess it should be possible to just pay for time on a machine from a grant to run such a computation from a Sage worksheet. But this would be well down the track. The reason I like this idea so much is that it makes formal mathematical papers interactive, which is probably a whole new paradigm which has been opened up in recent months by the availability of the Sage Notebook. It's clear a number of people have been thinking about doing something like this for a while, and I now claim that I am too excited about such a prospect. This could really make getting credit for work on algorithms and implementations feasible. Were there a proper international board of editors, highly respected in my field, available to organise formal refereeing of my paper through such a mechanism, I would be highly motivated to develop new algorithms, implement them, get them accepted into MPIR, FLINT, Sage, etc, write a paper on the algorithm using the Sage Notebook, including examples and/or benchmarks, and submit to the SagePapers archive. I'd then get credit for what I am not currently getting formal academic credit. And it would be fun too!! Hell, I'd even be inclined to write purely theoretical papers and illustrate them with examples using Sage and write the paper in the Sage Notebook and submit those too. And I see no reason why I wouldn't write review papers and expositions too, on areas of computational interest with which I am intimately familiar. As a first step towards something like this, setting up a website with a list of surveys/expositions which have been refereed and attached to a single Sage server, just as proof of principle would be an easy first step. Essentially the infrastucture for this already exists. We'd just need a distinguished editorship and some referees and submissions. Bill. On 12 Apr, 01:17, Rob Beezer <goo...@beezer.cotse.net> wrote: > Hi Bill, > > Grab the worksheet at > > http://buzzard.ups.edu/sage/sage-group-theory-primer.sws > > for an example of an interactive SagePaper (tm). There should even be > an @interact cell that will build a nicely formatted table of the > subgroups of a finite cyclic group, given the order of the group. > > On Apr 11, 4:51 pm, Bill Hart <goodwillh...@googlemail.com> wrote: > > > Another suggestion. Has anyone thought about having SagePapers TM, --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
