Active2 Dev Meeting Log, 15 May 2002

Loggfil \xF6ppnad vid: 02-05-15 20.59.29

*: Du har g\xE5tt med i kanalen

Anton: hi all

dizzy_wrk: Hello. When's the meeting supposed to start?

gek: got me

*: gek har opat Anton

*: gek har opat dizzy_wrk

*: gek har opat keysjrw

Anton: in 2 minutes smile

Anton: keith and scott can't make it but mike should be here any minute..

* Anton fetches some food..

*: mike (~mike@ har just g\xE5tt med i kanalen

mike: hi everyone; i don't have much time today frown, sad smile

Anton: nice seeing you anyway, mike smile

Anton: maybe we'll just sum up a bit where we are then and try to finish this meeting rather soon..?

dizzy_wrk: Sure. I s'pose I should point something out though -- I don't really know what this meeting's about. I don't have much to do with indymedia, but I've been working as a sysadmin for a couple of years on various platforms, so figure I can just poke my nose in where I can help.

mike: okay. i really haven't done much hacking this week on active2; i've been installing new hardware and latest debian on my system

mike: have introductions been made?

Anton: no

Anton: i'll starrt?

dizzy_wrk: Sure.

mike: i'm mike warren from calgary (alberta)

*: netd ( har just g\xE5tt med i kanalen

Anton: ok, i'm Anton from Gothenburg, sweden

dizzy_wrk: I'm Damian Gerow from Cambridge, Ontario.

netd: hoi! sorry for being late! I'm ingo from bielefeld, germany

*: mike,: No such nick/channel

*: Anton har opat netd

keysjrw: not from me, I'm James in Atlanta, want to start playing with everything in anticiaption of eventually getting atlanta off stallman, and agreed to try to make active work on RH7.1

*: Anton har opat mike

mike: any additions to the agenda?


Anton: oh, thats just me being eager to do some GUI stuff smile

Anton: sorry if i stress things wink

mike: that's fine; gui-work will be quite useful

mike: probably it will influence somewhat how the backend works...

dizzy_wrk: Is there a copy of active I can grab from somewhere?

Anton: dissy: just Active 1

mike: dizzy_wrk we're discussing development of active2

dizzy_wrk: smile I figured, given the name of the channel. So is there a copy of active2 anywhere? CVS?

mike: dizzy_wrk no, not yet

Anton: mike: sure, that's how I see it too.. more things will be needed to agree upon when we see an embryo to a GUI..

mike: i know where sf-active can be found, but that's probably not useful wink

dizzy_wrk: Okay, well continue with the meeting -- I'll pester people afterwards...

netd: mike: did you check out webware/cheetah? I'm amazed its a week gone already, time is flying by

Anton: however, a list of what we have agreed upon and must/should/could start developing might be good.. there are just some basic metting notes as of today in the wiki, right?

mike: Anton: as far as insertions go (agenda #1), we should hopefully have some simple-backend stuff ready to go in not too long

*: meta (irc@ har just g\xE5tt med i kanalen

mike: Anton: yes

mike: Anton yes, some more concrete documents in the wiki would be useful

* netd just did a s light agenda update

mike: once we decide what sort of framewrok is best to develop under, we could do this

mike: that is, if we're using mod_python vs. WebWare it might be somewhat different...

mike: actually, i guess we can do a lot of this now

mike: we have decided a number of things...

*: Anton har opat meta

mike: we should probably update the todo-list

meta: sorry for being late -- i'm in seattle with vic-keith

mike: meta NP; who are you?

*: meta \xE4r nu k\xE4nd som scott-van

scott-van: sorry

Anton: ok, #1 is done for now, right? we wait for a framework (is that it)?

*: Anton har \xE4ndrat \xE4mne till ""

mike: Anton first insertions can occur after there's a minimal backend

mike: so, yes

Anton: #2 (templates etc): has anyone looked more at this?

mike: no, not yet

mike: not until i get python 2.0, which won't be until i get a newer debain installed

mike: frown, sad smile

Anton: ok..

Anton: ingo?

* netd volunteers to work on the template part, if there's interest

mike: what is the status of ender? is it primary now?

scott-van: mike: apt-get dist-upgrade

mike: that is, do we now hack on hiro instead?

Anton: netd: i'd say great!

mike: scott-van yes, well, i first have to un-screw-up my hardware wink (new harddrive)

scott-van: (as keith) ender is about to become the primary..

*: scott-van \xE4r nu k\xE4nd som vic_keith

vic_keith: we'll start rebuiling hiro as the dev box (probably in the next 2 wks)

netd: mike: is there something I could look at already?

Anton: keith: can you put that info in the wiki somewhere too?

vic_keith: anton: what info? projected timelines for the dev server? i can keep the list updated.

Anton: yes, ok.

vic_keith: netd: it might be useful to look at how sf-active does templates

*: deanna ( har just g\xE5tt med i kanalen

Anton: i'd like a templating system where you name tags in some way but assign the tags a function in another file

Anton: not directly in the HTML

Anton: (as Chettah seems to do?)

netd: keith: as far as I can tell, its just keyword replacement. the syntax didn't look very intuitive for me. certainly looks error-prone. otoh, only the 'redstone' subdir seems to have some kind of templating at all

*: rabble ( har just g\xE5tt med i kanalen

vic_keith: is there a standard templating system/syntax out there? i guess cheetah but i haven't checked it out yet..

rabble: hey

netd: keith: I don't think there is such a thing as a "standard" templating syntax

rabble: cheetah is the best i think for python

Anton: rabble: do you have any experience with cheetah?

rabble: not much, but kellan spent a while playing with it and it sounds better than any of the zope connected stuff

Anton: ok

rabble: it has a lot of the really good features in template toolkit

*: rabble \xE4r nu k\xE4nd som kellan

* Anton points out the link just FYI

kellan: cheetah feels alot like velocity or webmacro is you've used those. its cacheing is pretty good. i think its basic syntax is nice. some of the more esoteric inheritance stuff should be stirred away from

kellan: but of all the python template systems, it was the best i tried

kellan: is a good source

netd: pong

* kellan and scott are talking in real space

Anton: ok, i guess #4 and #3 are rather similar: "Perhaps a list of things to accomplish, to give new people a place to start?" and "current state of the code"?

Anton: what to do about them?

dizzy_wrk: I added the fourth...

Anton: kellan: say hello from us all smile

dizzy_wrk: I was under the impression that #3 was like "Well, on this project we have this and this and this to do,

dizzy_wrk: so I was thinking along the lines of a general todo list. But it looks like that's being covered.

vic_keith: some to-dos came out of the last meeting as i recall

Anton: well.. i guess we agree we have to do it soon.. maybe just update with some general info too? (and the current state of coding)?

netd: afk

mike: sorry guys; had to step away

Anton: there's a lot of list/irc noise in the todo.. should be somewhere else IMHO

netd: back

dizzy_wrk: Agreed.

mike: anton

Anton: yes?

vic_keith: anton: good idea... maybe add todos to the meeting summaries on the wiki

*: mike_ (~mike@ har just g\xE5tt med i kanalen

mike_: stupid ssh

*: mike_ har l\xE4mnat IRC (Client Quit)

mike: okay, that's better

* netd is dizzy. gotta go

mike: cheers, netd

mike: how about

Anton: i see as a place to post coding/dev status and to post todo's smile

*: netd har l\xE4mnat IRC (Client Exit: Leaving)

Anton: wink

mike: if people can summarize what they're doing/have done then we can check out what works well, etc.

mike: maybe we should also think about general goals of the templating: we want something dead-simple for designers to use, which is a priority, IMO

Anton: maybe we all could start replacing all those maillist links with regular wiki info? i can start with some..

mike: it also needs to be flexible for designers

mike: Anton yes, having text directly in the wiki would be useful

vic_keith: mike: internationalisation is a priority as well.

Anton: mike: if i would suggest a "want" list, i'd say something like tag based \xE0 la Cheetah but with no SELECTs or $for $every $hit in it.. just something like %user% or something.. the $users variable should be set in another .py? file with assigned SQL statement..

mike: vic_keith yes, good point

Anton: the %user% would just correspond to the user field in the db

mike: Anton ideally, the template would render more-or-less nicely in a regular web client without the templates being grokked

mike: that is, a designer could write the template code such that they can see (in the above example) a table with a single row

mike: in a normal web browser

Anton: yes, exactly

mike: Anton we can't depend on SQL, due to the backend, though

mike: but yes, we need a way to embed backend calls

keysjrw: why not?

Anton: ok.. i see.

mike: keysjrw the backends might not be databases; they might be freenet

mike: (for example)

keysjrw: k

vic_keith: i don't feel like i can really comment on templates until i do some more homework (i.e. check out cheetah)

mike: vic_keith yes, me neither. sounds like we agree on the general goals, at least

vic_keith: mike: yup

mike: such goals will be useful to evaluate cheetah (or others)

Anton: true

vic_keith: so the goals could be summarized as:

vic_keith: ease of use

vic_keith: flexible

vic_keith: supports it8n

vic_keith: (speedy?)


Anton: and preferably free of coding language (other than html)

mike: yes, i agree. must-haves: 1. easy 2. flexible 3. I18N

mike: nice-to-have: 1. fast 2. renders nicely without template-expansion

mike: Anton yes, that sounds good too

mike: Anton although that would decrease flexibility

Anton: hm.. yeah i can see what you mean..

mike: Anton ideally: could be used without resorting to programming, but might be nice to have the option...

mike: that is, it's not hard to imagine a case where writing a snipped of Python would save much grief...

Anton: hm.. guess we end up with Cheetah then smile not my headache if you ask me wink just that we might sthrive to avoid it whenever possible, to place as much functional code in one place as possible..

mike: Anton yes, I agree that it would be good to keep code out of the templates

Anton: "functional code" is other-than-gui code wink

mike: maybe it would be best if this was a hard requirement

mike: smile

vic_keith: heh

mike: anyone else? should we reject templates which allow inline code?

dizzy_wrk: Out of curiousity, what would be the reasons for rejecting inline code?

Anton: from <TABLE>#for $client in $clients<TR><TD>$client.surname, $client.firstname</TD>

vic_keith: mike: that would give us the most long-term flexibility... but might make things unnecessarily complicated in the short term

Anton: maybe this does not infringe on my way of putting it? the vars can be set outside of the templates, right?

mike: dizzy_wrk makes lives of designers easier; keeps all the ``code'' in the server

mike: Anton don't know, but presumably

vic_keith: i'd like to look at how cheetah works before setting any hard reqs.

* vic_keith apologizes for not keeping up

Anton: ok, we might leave the (somewhat theoretical) template discussions for now then?

mike: Anton yes, let's look at cheetah (and others, hopefully) this week and then probably stronger opinions will emerge wink

mike: bottom line, though: there will be ``programming'' that the designers will have to use

mike: whether that is straight python or -type things is more the question...

* Anton wishes he had a CD-RW to burn that OS X image and try out some Apache/FreeNet/OpenSource stuff..

mike: well, sort of. wink

vic_keith: mike: heh.. are there any web designers who don't know any "programming" ?

Anton: i do know some php stuff but unfortunately not the "general" programming stuff (arrays, sets etc..)

mike: vic_keith yeah, probably not; javascript and stuf...

vic_keith: hmm.. actually there probably are

mike: Anton i know the way i design stuff usually goes like this: make nice ``example'' design, then chop out the ``example'' bits and put in my pseudo-template python code instead

mike: probably i'm just weird, though wink being a programmer primarily

vic_keith: mike: that sounds right.

mike: so ``ease of use'' for templates == ``easy to make example-tables into real-tables-with-real-data'', right?

mike: so if i had Some Name in my example, it should be easy to change ``Some Name'' to $ or whatever...

Anton: yup smile i'm familiar with adding and $fields and %fields% etc in my HTML code and has the tools for handling it wysiwyg too

Anton: however not "live" \xE0 la ColdFusion bridges etc..

mike: okay

vic_keith: mike: right. i don't have much experience with templates but that sounds right..

mike: we didn't discuss #3 yet, right? i've got to take off RealSoonNow

mike: the stuff i've done is in the Hacking node at the wiki...

vic_keith: cool

vic_keith: will take a look

vic_keith: will do some hacking myself once the whole ender/hiro fiasco has calmed down

Anton: mike: i propose we continue add stuff to your Hacking node whenever we encounter problems, tests template systems or produce code snippets

vic_keith: okay. anything else?

vic_keith: shall we call it a day? my blood sugar is dropping and i need some lunch smile

* Anton has nothing to add..

* dizzy_wrk has a couple of questions, if someone wants to stick around?

* Anton can stay, but maybe not answer wink

dizzy_wrk: Don't worry, they're just general questions. I take it the meeting's over then?

*: rabble ( har just g\xE5tt med i kanalen

vic_keith: kay, i think we're done.

vic_keith: dizzy_wrk: fire away.

*: deanna har l\xE4mnat kanalen

dizzy_wrk: Okay .. And remember, I said general.

vic_keith: ps. did anyone log the whole meeting?

dizzy_wrk: 1- Is there a mailing list for active2 development?

dizzy_wrk: vic_keith: I can copy it from my screen if you like...

vic_keith: yes:

Anton: keith: i log

vic_keith: anton: cool. could you put it on the wiki?

Anton: sure. not in a fancy way though..

Loggfil st\xE4ngd vid: 02-05-15 22.39.02
Topic revision: r2 - 21 Mar 2004, PaulWise
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback