Well, everything's fine, baby. For the past week I've been at work, errr, doing work stuff. It's actually quite challenging: we're completely building a new site from scratch (it's based on OpenFrame of course) and some of the bits are trickier than we had thought they'd be. It hasn't helped that James has been away ;-). We're writing lots more tests as we iron out the bugs, though, and I hope we'll be churning out the code soon.
For Christmas, my sister gave my father and I Japanese cookery courses for the next ten weeks. It's called Simple Japanese and it's at Kensington and Chelsea College. I've previously done really stressful cookery courses at Leith's. Simply Japanese is much, well, simpler. The teacher knows a lot and the other students are housewives, so it's all quite fun. She demonstrates and then we get to play in the kitchen. The dishes are simple (we made Kimpira Carrots, for example), but it's the techniques that we'll pick up which will be really useful (although ceramic knives look good too). Of course, I do have to get up early on Saturdays...
Last night I went to Aubergine a wonderful (if slightly expensive) restaurant on Park Walk. The dishes were great, the wine wonderful, and the company super. Of course, I really think Gordon Ramsay is the best restaurant in London. This somehow lead to a small discussion on #london.pm on what exactly British food consists of. Puddings? Suet dishes? Curries?
So, has anyone thought of coding a WWW::UsePerl::Journal so that I can grab my journal entries and put them on my website? Which reminds me: I've finally updated my website (it now runs under OpenFrame) and it has recent holiday photos and 100,478 recipes.
Took me a long time to type them all in...
Re:that reminds me
acme on 2002-01-17T09:57:57
Arrgh, good point. First thing next week, definitely. You can also come and talk to us on #openframe on irc.rhizomatic.net.Re:that reminds me
acme on 2002-01-21T10:47:28
http://www.astray.com/mailman/listinfo/openframeRe:that reminds me
lachoy on 2002-01-21T13:19:24
I'm on it like white on rice! Thanks for the ACK.
Re:sorta
Matts on 2002-01-17T10:24:33
Please.... no SOAP/XMLRPC...
How about GET/user/matts/journal with Accept: text/xml, and it returns an XML representation. And POST to /user/matts/journal will put a new journal entry up, accepting both HTML and XML (and maybe even POD!) in the Content-Type field.
HTTP fucking rocks, I wish people would stop using SOAP and XMLRPC when it's just plain not needed (not implying that's your intention, pudge, just most people see web services == SOAP). Of course I have a talk about that:-) Re:sorta
pudge on 2002-01-17T14:58:47
HTTP fucking rocks, I wish people would stop using SOAP and XMLRPC when it's just plain not needed
SOAP and XML-RPC *are* HTTP.
:p
I agree with your sentiment (when people started talking about this "new" thing called web services, I had similarly virulent reactions), but the key here is interoperability. More people can play this way. Further, this isn't just for use Perl, it is for Slash. I likely will use SOAP or XML-RPC or both, because I see no reason at all not too, and because it gives me the greatest immediate interoperability with the most systems, and it makes it easier for other web sites using Slash to hook in.
If you have specific reasons why you don't want to use SOAP or XML-RPC, that's fine, but "because it is not needed" isn't good enough in this case.
Re:sorta
Matts on 2002-01-17T15:20:26
Well I could point you at my slides, but the most important points are:
1. If you can do SOAP over HTTP you can do plain HTTP. It's not that hard, in fact I submit it's easier for the majority of languages (perhaps Perl excepted).
2. SOAP provides you with zero logging unless you do it yourself. Whereas with straight HTTP the server logs for you. If you use SOAP at http://use.perl.org/soap, then you might log 2000 calls to that URI a day, but you have no idea how they break down. That's bad, m'kay?
3. No XML parser required. No SOAP processor required.
4. HTTP is cachable at proxies and gateways. Never underestimate the value of caching.
For more information see the REST stuff.Re:sorta
pudge on 2002-01-17T15:36:43
1. Yes, I didn't say it was hard. However, I wouldn't say it is easier. Construct a POST request and handle authentication and send it, or call a method? While for *me* it is easier to do a simple HTTP request with all that, for most people it is not. Plus, there are already many tools and APIs out there for dealing with journals via XML-RPC/SOAP web services. For this growing group, hooking into Slash journals would be entirely trivial, if the APIs were compatible.
2. Slash already handles its own logging.
3. *shrug*.
4. Most of this data cannot reasonably be cached. It will be used primarily for creating/modifying/deleting.Re:sorta
russell on 2002-01-24T15:14:56
In the meantime, I've written WWW::UsePerl::Journal. Grab it from CPAN. Read only at the moment though...
