I'm just trying to think of what "new" could be, given that there's already some pretty lofty goals being aimed for in eventual development
aCiD2
right
ruaok
yeah, the key is that we need to have "new" in the sense of a "new start".
which means some less glorious work up ahead.
i18n is important. and now is the time. :)
nikki
and it makes me happy!
BrianFreudf
which is good :)
ruaok: totally different topic, but in terms of optimizing, have you ever considered turning on minification for the HTML that gets fed out?
nikki
minification?
ruaok
that and gzip.
BrianFreudf
yeah, hadn't checked, but I was guessing gzip wasn't on
aCiD2
BrianFreudf: As in stripping extra whitespace? could be tricky for how fragile the html is atm
ruaok
but what is amazing about all this is that since the RDF webservice traffic is declining and the XML increasing, we're constant on our bandwidth usage.
nikki
gzip makes more sense to me than stripping whitespace, html is terribly repetitive
BrianFreudf
aCiD2: I think it looks more fragile than it actually is - I've pretty much been post-shredding the site, and there's only one place I've found that's "touchy"
aCiD2
ah, ok
BrianFreudf
(the amazon text and image on releases - funky things happen there, I think because it's not in its own div)
ruaok
ok, I'm off to ETech in San Diego for some serious smoozing on MB's behalf.
BrianFreudf
have fun :)
nikki
have fun
ruaok
I might pop in occaisonally, but don't expect me back til late thursday.
thanks!
ta ta!
ruaok has quit
BrianFreudf
c ya
aCiD2: yeah, though, if I could just wave a wand and change all of two things, I'd 1) add mbid-support universally, and 2) add ID attributes to all elements
aCiD2
all elements?
BrianFreudf
yeah
crawling through 5 levels of nested tables with gEBTN is a total pain in the current site, when a single short gEBId would be so much cleaner and less fragile
aCiD2
true
nikki
if I had my way, I'd just strip tons of stuff out
but then again, I'm a minimalist
BrianFreudf
actually, it's a totally crazy idea, but I've been thinking, what if you broke the site into 3 divs. One for each menu, one for the "main view".
Then just have each menu as its own HTML page, the entire body contents of which you add in to the 2 divs with js after the real page loads
aCiD2
nikki: tons can be stripped out
but then I'm also a minimalist :P
nikki
hehe
BrianFreudf
then the menus a) come out of the mess they currently make in the HTML, and b) are cacheable
aCiD2
BrianFreudf: not liking that, you have to consider non-js users
BrianFreudf
sure, but you could always make it so the first page fed it always "vanilla", and if js is enabled on that browser, it gets a different version of the page which does that, else it just gets the vanilla version
UserErr0r has quit
UserErr0r joined the channel
sending the menu HTML over and over and over on every single page though has to be quite wasteful
aCiD2
yes, but comprosing navigation for non-js users is pretty much a no-no for me as for as usability goes...
BrianFreudf
that's why I say, make it so the menus are there in a vanilla version, but have a js-check on that page - if it detects js turned on, it adds a "&js=on" or something to links, so then the server knows it can feed the much smaller js-versions of the pages
nikki
hey, do you guys know much about sessions?
BrianFreudf
depends on the question :)
aCiD2 knows a bit
nikki
just in general, I don't really have any idea of how they work
simon tried explaining them when I complained about the sheer number of cookies the advanced edit search creates
but I forgot it all :(
aCiD2
cookie stored on client side holds some type of uuid, looked up on server side to fetch data
nikki
ah
BrianFreudf
basically a variable storing your info, so the server doesn't keep forgetting who you are or what you're doing
rpedro joined the channel
UserErr0r has quit
UserErr0r joined the channel
aCiD2
ok, cya tomorrow
time to watch more lost :P
Kerensky97 has left the channel
aCiD2 has left the channel
imoon joined the channel
chefkoch_ joined the channel
chefkoch has quit
VxJasonxV`
VxJasonxV` is now known as VxJasonxV
imoon has left the channel
mudcrow has quit
stochastic has quit
stochasticism joined the channel
BrianG has quit
Amblin- has quit
Amblin- joined the channel
futilius joined the channel
RodBegbie
RodBegbie has joined #MusicBrainz
futilius
hi
RodBegbie has quit
I have a special edition CD that has a bonus track that the regular version doesn't
how should I enter it? neither is currently in the database
Earliest version/later versioning is turning out to be far more trouble than it's worth...
rpedro has quit
rpedro joined the channel
srotta
pbryan: :P
pbryan: I guess I threw a few no votes your way.
pbryan: But look on the bright side, I didn't touch the "remaster"/"earliest release" controversy.
outsidecontext joined the channel
pbryan
Heh.
No problem.
I've canceled my remaining open edits until I get a better handle. The source of this information (booklets) is clearly incorrect and/or the releases I am choosing are not the originals to base it on.
mudcrow joined the channel
Now, to really cleanup the mess, I'm thinking I should undo the edits that did get accepted.
Thoughts?
srotta
pbryan: The Queen edits, or are there other ones as well?
pbryan
Queen and Essential Dave Brubeck
I tried 4 discs as a pilot.
Neither turns out to be acceptable without significant amount of research.
Problem is, a bunch of my edits got in now. :)
Any AutoEditors want to make my life easy? ;)
srotta
The accepted Queen edits I checked, they should be ok. The only ones with problems were the ones I noed.
pbryan
Okay, well they should be toast (soon, when cancellations kick in).
srotta
Ok, there may be some that are AR'd on album when a single was released before the album, but that's nit-picking - I usually add those ARs to albums myself (if the single and album versions are the same, that is).
pbryan
Sure, but if the track times are way off, then it can't be the same version.
srotta
Yeah.
pbryan
Thanks for catching that.
srotta
At least One Vision might be an original edit to that compilation, but I can't be sure since I didn't find any complete version information about it.
pbryan
I truly can't be sure unless I have the collection on which to compare the compilation tracks.
I think it's best to leave it to those who have artist collections to do baseline comparisons of.
"I read it in a booklet" has turned out to be insufficient.
srotta
Yeah. That Classic Queen had some information in Wikipedia, and Wikipedia in general has some information on different available versions, but not on all tracks.
Yep, too often booklets contain inadequate information. Many compilations (for all artists) contain single versions of tracks that differ from the album versions, but you don't often see that mentioned in covers.
pbryan
An another topic, I'm finding many releases are released under multiple labels. One disc has three labels: Wide Open, Monument, Columbia. Thus far, I've been relying on what label is most prominent on the spine and cover to pick one. Question #1: Is this the best approach. Question #2: Any enhancement requests outstanding to allow a release instance to tag multiple labels?
warp
pbryan: for choosing labels, I mostly rely on the catalog number matching other catalog numbers of one of the labels.
pbryan: but i tend to listen to music which is released by labels which use recognizable catalog numbers... it's not very useful if it's just a barcode in disguise ;)
srotta
I usually pick the "smallest" one, but on the other hand I usually deal with labels I know.
pbryan
If it's ambiguous, best not to specify?
srotta
So I use Herodes for Sir Elwood releases instead of EMI, even though both labels are on the backcover and there's no distinct cat#.
pbryan
Because Herodes is more distinct ... granular (less releases under them than parent giant EMI)?
srotta
Some people add a release event for all the labels. I'm not too fond of multiplying a single release event.
pbryan: Right.
pbryan
Alrighty. That's been in the same vein as I've been following.
Just not definitive though, unless I can tell what catalog the catalog number actually ties to.
outsidecontext has quit
warp
pbryan: it's definitely better to have whichever label from the backcover than not entering one at all.
pbryan
Alright. I'll keep that in mind.
Anyone considering allowing multiple to be specified per release instance?
warp
multiple labels?
srotta
I don't know. Not a bad idea, at least it would be better than guesswork 8)
outsidecontext joined the channel
mudcrow
multiple labels = good idea, but would result in a huge mess when most people cant seem to identify one label, let alone multiple labels
warp
a single release event is released by a single label normally, it's just not always clear which one. allowing multiple labels is not a solution to that problem.
mudcrow
i saw MCPS as a label the other day with C78 as the cat number. MCPS is the royalties collection agency, C78 = copyright 1978 :/
srotta
I'd say finding the correct label on some releases is damn near impossible. And since the release mentions multiple labels, it might be a good idea to allow people add those multiple labels. Even better if you knew their roles in the process, but that's even harder to realize :P