delirioussitedump back up again (for the time being) The delirioussitedump is back! The dump was performed on Sunday, up through entry #49242. I'll probably start running a daily update next time I have a few minutes to work on this, but for now, that's what's there, hand updated only. Yes, it's officially in YAML now, and it's bz2 compressed. The compressed version is some 2 and a half megs, whereas the uncompressed file is 13 megs (!). Substitute .xml for .yml if you want my ad-hoc XML instead of my ad-hoc YAML version. The uncompressed files are also available, at least for now, although I would vastly prefer that you download the compressed version. See also One of these years I'll also publish some statistics, most popular URLs, etc. (I actually already have this information extracted, just not in a convenient format.)

Just for a quick review, here is the dump I got by scraping the site the other night (and/or day) This is temporary, and will disappear. Soon. It's just for you to comment on before I do the next round (tentatively, I'd do this weekly). Please review the details of the YAML formatting in particular, if you think you know how it should look. *** This is a 770,167-byte file on a slow server [now up to 1,3 Mb]. Don't click if you're impatient. I deleted entry #247 altogether. Other than that, the data is as I got it off the server. I obtained the data by going through the /entry/nnn for each nnn up to number 2519 [now 4259], which for some reason I believe was the highest entry number at the time. Then I extracted the actual data out of the web page for each entry, and converted it to (some sort of) YAML. The /entry/ page doesn't show the modification time, so I don't have that. I have reserved a field for it in the format nevertheless. I also have this in something resembling XML, of course without any formal DTD or anything. Let me know if you think getting that would be more useful (instead, or as well). For the context impaired, the "and/or day" refers to time zone confusion.

Please only use it for topics related to dumping the de.lirio.us database Also, I doubt that "chmod 777" is ever the correct solution for anything related to hosting web pages. But other than that, I guess it's good to have a reference for setting up your own rubric installation.

For the time being, I believe it's completely undocumented Wishlist (1): get this on de.lirio.us too Wishlist (2): include each entry's number (id=) in the dump Wishlist (3): document at least the existence of this facility

While steve@de.lirio.us is broken ... Steve Mallett's "about" page at fooworks.com includes contact information, in case you need to get a hold of him urgently regarding something at de.lirio.us. I exchanged email with him last week about the problems we are currently seeing ("internal server error" whenever you try to do anything even remotely heavy, such as, oh, say, view your own entries ... duh!) and apparently he is working on solving them. While we are holding our breaths, I have found that the RSS feeds seem to work even when you otherwise can't seem to get anything to work, although of course, they are only updated with some delay, and there's no way to get to a page where you can edit an entry. (I frequently need to go back to an entry I just added ... but I guess I should just learn not to click "save" too early.) If somebody has a pressing need to switch to a different provider, I may be able to help you collect all your entries (except the @private ones of course) -- I still have the simple mirror script which I have used for the "delirioussitedump" (see this tag) although of course that too has been working less than perfectly during the last few weeks. Get in touch or ... oh heck, I'll just make it available again at the previous location. I'll post an announcement in "delirioussitedump blog erablog" when it's there. Update 2006-02-06: click on the "delirioussitedump" tag above to see the announcement.

In case you're wondering, my server is *still* offline They finally managed to get an ADSL to my new house, a few weeks ago, but ... by their policy, you can't run a server on that. The local not-a-monopoly are apparently stalling in every possible way, but will eventually have to give in, and then my old "pro" ADSL will be moved to my new address. Updated ETA: July 4th (happy ... naw, hopefully possibly slightly less neurotic birthday, USA). Update July 13: They're still, errrr, incompetent. Hold your breath for another week or two, maybe.

ETA for the new ADSL connection is ... May 17th Happy birthday Norway. Incredible, three weeks just to connect a piece of wire to the house?

If that's not what you want, can you explain what you do want? I have a cron job which pulls down the entire database, but it won't get me any @private entries. If you want only the (non-@private) entries you created, I can extract those into a separate file for you, but I can't promise very timely delivery. If you want to pull down data yourself, I can certainly give you a copy of my scripts -- but then, if you are capable of figuring out how to run them properly, you are probably also capable of creating your own equivalent scripts. To contact me, try the email address ee are a (that's three lowercase letters) at ih kay ih dot eff ih ... but please don't store it in your address book if you are using the (lack of) operating system which is most prone to email worms and viruses.

Since this data is kept someplace, and might help readers grok what they're seeing ... And of course, so that I can get it into the site dump. You can see the last modification date of your own entries by opening the (edit) link for them, but obviously, you can't do this with other people's entries.

Getting the site dump would be expedited by having a way to find out the index of the last entry The way I do it now is just fetch the cover page, extract the number of entries from the footer, and assume that's also the number of the last entry. I know this is incorrect because the number of deleted entries is subtracted; I could adjust by adding the number of deleted entries I already have from the previous dump, and get closer to the truth, but the ideal would be to get the index of the latest entry, no more and no less. How about a parameter to force the display of "(body)" links for all entries?