Placing a foot into the other camp
More migration seems to be occurring, so it's probably about time I got round to doing this: I'm now simont.dreamwidth.org as well as simont.livejournal.com. I have no plans to deactivate my LJ or stop reading stuff via it; I expect to cross-
There will doubtless be annoying teething problems as I sort out this dual presence and integrate it with the rest of my setup. In fact I've already made one cock-

no subject
no subject
How confusing for anyone not reading both!
no subject
no subject
no subject
Irritatingly DW doesn't seem to be able to send me notifications of my *own* comments here, only of replies to them. Although I should probably go double check that's still true
no subject
no subject
no subject
no subject
no subject
Page layout: actually, I sorted that out way back in 2009 when I set up my OpenID presence, and did the porting work to adapt my custom style to the (small) changes in the Dreamwidth version of S2. So this time round it was no trouble at all – I just pasted the layout code I already had into this account, and it worked just as well here as there.
(Though I notice there are a couple of things wrong with it, most immediately the absence of the standard Dreamwidth 'you are logged in as foo, handy links and stuff' bar at the top. I may have to work out how to get that to show up.)
Welcome: thank you! :-)
no subject
The layout porting thing, I should have guessed you'd find it pretty straightforward! I keep getting people asking me how to do it, and I say, oh, you just change these three variable names and these two functions, and they complain that's way too complicated. But we did try not to break S2 too much when we were modernizing and rationalizing it (and that was a dev task that I was pretty heavily involved in, so I'm somewhat proud of it).
Custom posting software, hm. What are you doing for backup / archiving? Because I used to use LJArchive, which has a Windows GUI, but it times out trying to back up the whole of my Dreamwidth journal. I think what I need is some kind of script I can run that will talk to DW's content exporter (Perl or Python or whatever more knowledgeable geeks convince me is suitable); do you have any ideas about that side of it?
no subject
Thanks; I hadn't spotted that. Unfortunately all the boxes are ticked and the navigation strip still isn't showing up for me; my guess is that that's because my S2 code is failing to call whatever method optionally inserts the navigation strip into the generated HTML. I should probably just go and find the docs and work out what I should have called and where. (Which of course will be easier now you've told me the proper name for it.)
Custom posting software, hm. What are you doing for backup / archiving?
Hmmm. Perhaps I should have said "horrible hacky script" instead of "custom software", to give you a better idea of its general level of organisation :-) I don't have a full-on client by any means; all I've got is one script which lets me write posts in a markup language I made up off the top of my head, and then formats them into HTML for LJ, into plain text for posting to a private journals newsgroup, and into colour-enhanced plain text for posting to Monochrome BBS (where I've maintained an online diary since before LJ and blogging were famous :-).
Archiving of my LJ is something to which I have so far not taken a particularly joined-up approach. The posts I make are archived in the form I post them on Monochrome, because I happened to already have the technology to do that. Comments on my posts are archived either in my email archive (which is why I like self-comment notifications) or in a big directory of HTML files I created one day by spidering my entire LJ (the idea being that the latter covers all the comments from before I started keeping those emails).
It would probably be a good idea to look into sensible content exporters, but it's a long way down my to-do list at the moment...
no subject
Page::print_control_strip(), as it turns out.no subject
no subject
backup/archive
However, I haven't tried to get either of them to speak to Dreamwidth so I can't say whether that would work or whether the API has changed too much. Worth a try, I suppose.
Re: backup/archive
I'll see if I can get the Python script you mention to work, cos I feel uncomfortable with half a year of DW only existing in the cloud!
Re: backup/archive
no subject
In the page you link to there I see a lot of configuration options for inserting a footer on the LJ side. But what controls the "xpost" footer on the DW side, as on
no subject
set text_meta_xpost = "Crossposts";to change what it's called in the footer. Changing how it displays beyond that, at a first glance looks impossible (it's in the so-called metadata, which is all called from the backend and you can't do anything with it by tweaking layout code.)no subject
no subject
no subject
I think I've set up my posting script and LJ crossposting properly now (though I'll know for sure the next time I make an actual post), but one thing I haven't got to work is the DW equivalent of the LJ "checkfriends" service. When I ask LJ if there's been an update on my friends page, I get this output from my polling script (which just formats the input and output parameters of the request as Python dictionaries):
Input: {'lastupdate': '2011-03-15 10:32:55', 'mode': 'checkfriends'}Output: {'lastupdate': '2011-03-15 10:32:55', 'interval': '120', 'success': 'OK', 'new': '0'}
But doing the same on Dreamwidth gets me this rather less helpful thing:
Input: {'lastupdate': '2011-03-15 10:32:55', 'mode': 'checkforupdates'}Output: {'lastupdate': '', 'interval': '36000', 'success': 'OK', 'new': '0'}
I've searched the Dreamwidth wiki and found the documentation for the method, which suggests to me that "lastupdate" should not be coming back as a blank string. (Admittedly I'm calling it via the LJ-style "flat" interface rather than the XML-RPC version, but it doesn't look as if that should affect anything more than the formatting of the wire data and the underlying set of fields ought to be the same either way.) Also, that poll interval is pretty unfriendly – I thought the purpose of this method was to give people an alternative to constantly reloading their friends/reading page, which it surely won't do if it demands you only call it once every ten hours.
So would you happen to know if this protocol feature is supposed to work on DW at the moment (or if it would start working if I got a paid account, perhaps), or failing that where else I might usefully look other than the docs I've found so far?
(I also couldn't find full documentation for the more sensible forms of challenge/response authentication. I can see the getchallenge request type, but I haven't found anything about how you encode a response given your challenge.)
no subject
However you're getting to the point where you're asking questions beyond the limits of my technical knowhow. You've picked up enough from the docs to replace
checkfriendswithcheckforupdates, which is the usual thing that goes wrong with porting stuff over from LJ. I have a feeling it's deliberate that it's set at an excessively long poll interval (and quite possibly disabled altogether) for free accounts (it used to be that way on LJ too, at least back in the day).The official Support place is http://www.dreamwidth.org/support/submit. But there aren't many volunteers active there with more technical knowledge than me. For this kind of properly technical question, you'd do better asking in
no subject
Hmm, now you mention it I do remember checkfriends being a bit iffy when my LJ account was a free one. (Which it hasn't been for years – I got given a permanent account free of charge in gratitude for PuTTY – so my memory of free account behaviour is somewhat hazy.)
I must say I never really understood that policy on LJ, though. I think it's a mistake to pitch checkfriends / checkforupdates as a premium service provided for the benefit of the user paying for it; its purpose is to reduce load on the LJ/DW servers, so it's a feature provided for the benefit of the site administrators, and hence it surely should be more beneficial to the site admins the more users it's provided to. In particular, if I can't use checkforupdates at two-minute intervals, then the obvious alternative is not to wait ten hours between checks but to reload my actual friends page every two minutes, which is more CPU-intensive due to having to run the S2 and lay out all the HTML – so by denying any user a working checkforupdates they're increasing the load on their server.
Still, that's not your problem, of course. And I suspect I may end up getting a paid account anyway (I want a network page, notifications for my own comments, and comment editing – LJ's gift of a permanent account has apparently had the effect of getting me accustomed to those features, and hence making me want to give DW money :-), in which case it's not my problem either...
no subject
For the record, I can now confirm this: I upgraded to a paid account this morning and suddenly checkforupdates is behaving sensibly.
I still don't see the sense in the policy, and it would have been nice to have an error message rather than getting back an unexplained blank string (I did get a nice error message when I tried to use checkfriends, which is how I knew to rename to checkforupdates), but at least the cause of the issue isn't in doubt any more :-)
no subject
no subject