30 Jul 2012 cdent   » (Master)


Created tsapp to display, in a very basic way, the functionality available via the tsapp tool that I've been creating over the last few days. It worked really well.

There are a few issues but for the most part the concept seems sound and the implementation simple enough that it doesn't make my ZOMG complexity! Run! reaction kick in.

One open question: Should it allow user registration and space creation? At first glance this seems a step too far, we'd want people to have understood what TiddlySpace is before using tsapp. But perhaps this is wrong, maybe tsapp is a vector for new (technically oriented) people to enter the ecosystem?

Another issue is how to deal with an app that is designed to be included or uses "extra" bags. Is special support for that needed? Or can that be worked around.

It may be useful to compare what was created and what was discussed. The discussion described the form of an app and the process of app creation. The form of an app is mostly the same between the two. The major difference is that at the moment tsapp has no facility for pushing to "optional additional bags". You can proxy to those bags, but when tsapp push is run the content goes to just one bag. It may be worth considering a way to map assets to different bags (presumably by pathname).

The discussed app does both push and pull. The created app only does push. Implementing pull assumes there are situations where the server is authoritative for content. Either the local dir or a git repo ought to be authoritative. Also GET is the easiest of the methods, so curl can be used in a pinch.

The discussed app planned "reliable algorithmic transformation" of paths to allow content both local and remote. The built app goes for a much simpler solution, that seems to cover most common cases. So, in fact the "overkill" solution was the simple one. Largely because WSGI is awesome.

Syndicated 2012-07-30 19:47:47 (Updated 2012-07-30 21:33:12) from cdent

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!