Slash 2.2.0 Released
If you meander over to Slashcode, you will notice that Slash 2.2.0 has been released. This is of course the website engine that runs Slashdot. The release has the message system, improved journal functions, new comment filters, and countless bug fixes. And of course a variety of optimizations that continue to make it possible to serve a quantity of pages that no other open source package like this can even touch :) Plus it's way easier to install. Now that we've got the Fry tree out of the way, its off to work on Zoidberg (which will include subscriptions, killfiles, and a few surprises)
Any chance of there being a DEB package of Slashcode? Does anyone know? Anyone working on this?
=U= "Just because you're not paranoid doesn't mean they aren't out to get you"
I mean really, we've been looking at the same interface for years. Not that there's anything wrong with the current appearance, it's just time for a change.
"If you love someone, set them free. If they come home, set them on fire." - George Carlin
I'd like to be able to put in an URL something like
5 &s tyle=light
...
http://slashdot.org/frontpage.pl?commentthresh=
and have it give me slashdot in 'light' format, with comments in the stories as 5 & over only.
The reason for this is that I want to get Slashdot on Avantgo, but obviously I have different viewing requierments on my Palm than I have on my desktop.
Is there any way of doing this with Slash 2.2?
I know there wasn't in the old Slash 1.x
I want to be able to set my preferences to include stories that only appear in certain sub-categories (Science and Ask Slashdot) on the main page as if they were full-fledged main stories.
Here is something that sort of irks me. My karma is capped, not that it is important or anything... but get this...
I post a comment with a score of one. Four moderators come along and think that the comment ought to be at a five. Three more moderators come along and they think that it should be at a two. I just got screwed out of three points, when I really should have stayed at 50.
50 + 4 = 50 (karma cap)
50 - 3 = 47
I just lost three points when in reality I should have gained one. This sucks.
When I give my indoctrination spiel, uh, er, "training lecture" on CVS to coworkers and lackeys, I usually hand down a list of requirements for log/commit messages and a ChangeLog entry.
Most of the time it's a variation on http://www.gnu.org/software/guile/changelogs/guile -changelogs_toc.html,
or sometimes the rules used for ChangeLogs in the GCC project, which I've found to be of immense value when tracking down changes.
Personally I can't stand changelogs that don't say a thing. It's just enough "open source" to look good, but not open enough to actually invite help.
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
How about comment editing capability??? There's nothing worse than posting a comment, even using preview, and realize you screwed it up somehow.
It might even be interesting to add a "previous version" capability. Just stick the message in some other dump table and have a different screen to dump them out.
The journals have editing capability, so I don't see why normal messages can't do the same.
Sometimes it's best to just let stupid people be stupid.
Four suggestions -- three of which should be easy and the fourth is harder:
I agree that the UI should default to use year indication in its default date format - without it is a mess when searching old articles.
Preferable using a universal understandable date format, which do not leaves you wonder if month or date are written first.
yyyy/MM/dd hh:mm, e.g. 2001/11/07 19:25
or
dd month yyyy hh:mm, e.g. November 7 2001 19:25
In favor: I've used and enjoyed Slashdot for a very long time. I'm not concerned about privacy issues involving my email address, so that's not a worry for me. I know that a lot of hard work has been done to keep this service running for me to enjoy, and I know that the upkeep costs a lot. I know that the reality of the web is different now than it was.
Convincing, but against: I, and all the other posters, experts, flamers, trolls and etc. are what make Slashdot even basically interesting. The stories alone I can get anywhere -- it's the posts that are semi-interesting. When I pay for a subscription to Salon, I'm paying to get content I enjoy. If I were to pay for Slashdot, even just to get ride of ads, I feel like I'd be paying for something I help make happen.
All that said: I'll pay for Slashdot. The reason is that, all philosophical problems aside, I know that economic realities are forcing this thing in. I'll miss the free Slashdot, but even a subscriber-friendly Slashdot is better than no Slashdot at all (what would I *do* with my days?).
In return, I'd really, really like to see a more intelligent basis for story selection. I *miss* that Slashdot from three years ago where the stories were mostly tech-oriented and not just another excuse to flame Katz or diss Microsoft. I want to see real efforts to improve the signal to noise ratio without stomping on unpopular views (like moderation tends to). Maybe it's not possible to go back to that, but I'd like to see some effort made to try.
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
The moment it costs more than $0.00 for any content on /. I for one will no longer be a reader of Slashdot. The next we know it VA software will pull up the leash, and slashcode will become proprietary
Nobody else has mentioned it, and it's my pet peeve, so I'll throw it in there -- I would love to have nice clean XHTML or XML that could be formatted with stylesheets (CSS or XSLT) on the client-side. Now that Mozilla is out there, this should be politically acceptable.
This could potentially reduce serverload quite a bit -- not only would you be spitting out far less bandwidth per page, but things like score filtering could be done on the clientside instead of requiring another roundtrip to the server.
You could even invent your own killfile, highlighting, light-mode, and score biasing schemes. Slashdot could use a default stylesheet, and then host user-submitted ones. Removing all the presentation goop would probably make NNTP/Gopher/whatever gateways easier to implement too. This would also have the positive aspect of pushing off most of the minor bitches back onto the userbase.
Whenever I hear the word 'Innovation', I reach for my pistol.
Reproduced for the terminaly lazy:
In order to make that usable, I'd have to pump my link depth to something like 4 in order to read the stories. Plus, for the first time in months, slashdot.org has stopped serving 403's to sync.avantgo.com, which basically killed it's usefulnes... (It was one of the first sites I tried to sync to my iPaq via AvantGo, and until today, everytime I tried, I'd get access denied errors reading it when I tried to sync.)
You are in a maze of twisty little relative jumps, all alike.
Hmm. I haven't seen a PHP package that would do nearly as well as Slash under heavy load. Kuro5hin is a perfect example - while they have similar hardware their site is usually slow and often painfully slow. And that's with a fraction of the traffic.
No, Thursday's out. How about never - is never good for you?
<a href="//slashdot.org/whatever.pl...">
So. w3m interprets this as http://slashdot.org/slashdot.org/whatever.pl. OK, it's probably valid html since it works correctly in Mozilla and friends. But still, I wonder why the fsck does such a convention exist in the first place. More importantly, what's the point of repeating the host name for local links?
--
The Cap is nigh. Time to get a fresh new account.
Why the continuing trend to offload debugging onto users who didn't ask for the 'improvement' in the first place? Why keep pushing code onto a hugely popular community site that only serves in Generalissimo Taco's war on trolls, dadaists, and the generally absurd? Why a gzip "filter"?? There are decades of research into fast algorithms for determining statistics on bodies of text -- any one of them, many public domain, would be an intelligent tool against crapflooders. But a 'compress and size check' line instead?? That's the worst kind of lax unfeeling code, that wields a brutal metric without regard for corollary damage. ("Rob code", I've heard it called, but "MS code" is more typifying of that style of program design.)
I generally don't rant. You are running a valuable site at no monetary cost to us. But /. continues to become a place where trenchant technical analysis is unwelcome, master geeks ignore the pablum, and Taco & the trolls continue their little war with the rest of us caught in the minefields that they lay. Every day I feel a greater desire for the /. of 1998. I say, bring back Chips and Dips.