The goal: XML-based import/export with easy UI

[Dave Winer]

I had lunch last week in NY with Matt Terenzio, and realized I didn't
have a mail list for him that made sense. I saw that as a bigger bug
than just not having a spot for Matt (he will certainly say something
snarky here).

There's a reason the OPML Editor has the same name as the format --
because it ties all of it together, at a user level and all developer
level. By "it" I mean my work. :-)

The last few days I've been working on something I've been working
toward for about a year now -- being able to move whole blogs around
with nothing but XML files. We've done this before, but this time
we're going to do it slowly, and really build systems on it. I hope.

The first part of the transition is already done. Radio2 has always
outputted backups of the whole user's site every time there's a
change, but it saved the files in .fttb format, which is only
understood by Frontier. No other app should try to process this
format. At the same time we were maintaining an RSS archive of all the
user's posts. So most of the data was already being output as XML.

The next step was to also export the contents of the users prefs,
stats and callbacks from config.radio2.users.

That is now being done, as of yesterday. If you look in the log of
Radio2, if you are using it, you'll see that once an hour it updates
userData.xml for all users who updated in that hour.

http://scripting.com/images/2011/11/21/backup.gif

Today I'm working on sucking in these files and then also sucking in
the archive of RSS data.

When this part is done, we will have a full and totally open import/
export for linkblogs.

Then the goal is to make the UI for import/export a total snap.

Dave

[Matthew Terenzio]

something snarky  . . . : p

Looking at future uses of this feature, it would tie well into the issue you often raise about the long term lifespan of the content we create.

So, for example, we could take the data and write a script to turn it into a static S3 based site to park for generations to come.

I'm sure you are way ahead of me on that.

[Dave Winer]

1. I've now got the script that sucks down the whole archive, so the concept works.

I have yet to prove that Radio2 will work with the sucked-down archive, but that comes next.

2. The content already is in an S3 static bucket. It starts there.

3. Yes, this is very much related to all the harping I do about future-safe archives. Radio2 incorporates all that I've learned so far in that.

Dave