Report problems, make suggestions, request features here.
If I set up a fossil repository when I switch server, is that fine?
Name:
Anonymous2013-09-02 18:06
Why can't you set one up here for now, then when you get the new server, you copy the partitions/files over?
Name:
Anonymous2013-09-02 18:15
>>2 There is a reason. Also, I'd rather not duplicate work. Since I never use Apache, getting random things to work with nginx takes a good amount of messing around with configs.
Name:
Anonymous2013-09-02 18:16
Is this a local problem or is the new /prog/ displayed smaller than the new one?
>>4 It's not just you, I set the font size a bit smaller.
Name:
Anonymous2013-09-02 18:24
>>4 We are using table cat's software for now. We are going to switch right? What should we switch to?
Also, I'm posting from tor. Would it be ok to leave tor unbanned? Pretty pretty please? I don't trust you the owners of the infrastructure leading to your server and/or the owners of the physical server machine with my ip address.
Name:
Admin2013-09-02 18:27
>>7 I'll leave it unbanned if it doesn't cause problems. I probably won't ever really ban IPs anyway, I'll just delete / filter / prevent spam using my EXPERT PROGRAMMER skills.
I plan to switch to my own textboard eventually. Though, there's also Kareha? Tablecat really isn't bad, though.
Name:
Anonymous2013-09-02 18:29
>>8 I'd rather stay with Tablecat. I like it a lot, and I don't see any reason to switch to Kareha.
Name:
Anonymous2013-09-02 18:30
>>8 Sounds good. It'll be fun to develop one in (scheme)
Name:
Admin2013-09-02 18:36
>>10 I actually wrote an interpreter to use with nginx because 1. I couldn't actually find a server-side Scheme thing anyway, and 2. the ones I could find in Lisp required you to use an entire web server / interpreter monstrosity. Whereas I'd rather just use nginx and then summon the interpreter through FastCGI.
I'm more or less done with the interpreter itself. It's more than fast enough to be used for something like this. I'll look at the FastCGI spec soon and write an interface for it, then we can put it all together and use it with the Scheme BBS.
>>11 Interpreters are fun to do as well. You probably could have just hooked in chicken. It's pretty easy to do stuff like that when you can just inline C, but I don't know much about nginx.
Name:
Anonymous2013-09-02 18:56
>>8 Can you make a hidden service? You know, it's funner to use Tor to browse a hidden service than to browse a clear site.
There's little point right now since I already registered the domain and everyone already knows the IP of the server.
Name:
Anonymous2013-09-02 21:35
>>13 None of the interpreters I looked at were extensible to the point that I needed. I have to integrate a FastCGI module into the interpreter if I don't want it to be horribly slow.
I'll release the code for everything when I'm done with it all.
Name:
Anonymous2013-09-02 21:44
>>16 Hmm, something tells me it might be possible to compile all the cgi to native code. I don't know if that's a sane way of doing it or not, but it seems doable.
Name:
Anonymous2013-09-02 21:48
>>17 Well, the point is not compiling it to native code. You can compile Scheme code into a pretty efficient representation for most purposes. The point is that if you don't write integrate a sort of FastCGI daemon into the interpreter (as some sort of extension), then every time that the page loads the interpreter needs to start, load the environment, load the script, set up the connections, etc, etc. And there's no possibility of caching anything if it's starting up from zero every time.
That's the real issue, and that's why I wrote my own interpreter and am writing my own FastCGI daemon for this. It would basically just do what php-fastcgi and perl-fastcgi do, it runs as a daemon on a port and nginx feeds it the script to run along with the HTTP parameters passed, and the daemon fetches the script from the disk / cache, evals it, and returns the body to nginx which then relays it to the client. Much better overall.
Name:
Anonymous2013-09-02 22:12
>>18 Ah, I understand the architecture now. Yeah, if you are keeping them separate like that, that is the best way. I noticed the same problem when I was looking at implementations to use for this. There wasn't anything out there like perl-fastcgi. There was mod-lisp, but it has more overhead of passing data through a local socket.
I was assuming something like compiling the whole thing, cgi and the web server, into one monolithic binary or a bundle of shared libraries.
Name:
Anonymous2013-09-02 22:20
>>19 That's annoying if you have to change or update stuff, though. If I wanted to go that route, I'd probably just use straight C. I'm more than comfortable enough in C to do that, but I think Scheme lends itself really well to server-side scripting and so I wanted to write this. As you said, there isn't really anything like this out there, which is a shame. So perhaps if I can pull it off well enough it might find many uses, and make Scheme a little more used in general.
There is a reason. Also, I'd rather not duplicate work. Since I never use Apache, getting random things to work with nginx takes a good amount of messing around with configs.
I am confused, shouldn't it just be a simple migration/copy of two jailed processes on the same server, with the switch of DNS zone? >>15 I see, so you are worried about the IP address being exposed, since you may have things on this server already; which maybe why you looking into running hidden services. Try Gnunet and/or Freenet. Tor if you want to risk it like you already have.
>>21 That isn't my repo. Don't worry about the reasons. I'll migrate this when needed. I've already half completed the FastCGI interface, so it's all fine, anyway.
>>22 I plan to set up a repo and put the code on it when I have it at a respectable level so you can all contribute if you like. That's what I was thinking, as well, with the security thing. Plus, it's always better to separate things and have one thing do its thing well, rather than have something try to handle everything at once.
Much more useful if the parts can be combined in whichever way.
What's with the timestamps? Doesn't seem to be UTC {-7, -5} (murrika times), UTC {-6, -4} (murrika summer time) nor UTC 0 (DA ISO STANDARD). So what timezone is it?
>>28 What if ``Don't bump'' just turned the name blue? I can't really think of another use for emails aside from having johhnyizcool1997@outlook.com be a tip-off for irony.
>>27 I'll get around to it. Self-signed because I'm not paying for the nonsense CA web of retardation.
>>28,29 I'm reluctant to actually change the board software itself. Perl is one of the few languages I've done very little with because I despise. It's also usually hard to read due to VALID PERL CODE. I'll look into it but if it takes me more than 5 minutes it's probably not happening. I'm also fairly busy with work so I don't have time to look after this 24/7, I figure I'll need to appoint a moderator to help me out eventually.
Name:
Admin2013-09-03 6:10
>>24 I'm not really sure why the time zone is like that since the server is in the US. Doesn't concern me much, really. >>25 Looks fine to me. If I change the size it might start messing with kopipe.
Seems like Tablecat just appends a number inside the post class (class="post [10][timestamp]") that it then parses again on read / index build to figure out which posts were sages or not. If 1 the post was a sage, otherwise it was a normal post.
>>39 I don't understand why it happens. I think Tablecat only allows one post for every 10 seconds no matter the IP. I'll change it now, I guess. Really weird.
>>40 Okay, I was under the impression that FLOOD_LIMIT was IP based, but it isn't, it just stats the time of the last log file write and compares that to the current time.
Looks like I'll have to implement IP based rate limiting myself. Right now there's a global limit of 1 post per second. I'm surprised we ran into the 1 post per 10 seconds limit already.
FUCKING PERL WHAT THE FUCK PERL PERL PERL
Name:
Anonymous2013-09-03 21:18
>>41 Dear god. That sounds worse than Shiitchan's flood detection.
>>42 Yes, this software has some... issues. Mind you it's pretty amazing what the Tablecat guy has done with how restricted he chose to be in implementing this (no database, just a series of files and an index)... but some of the stuff makes me question his sanity, such as building the front page by parsing the HTML with regex, etc.
>>44 Same. I think it's pretty neat in general. I'm not sure how well this implementation will scale over time, though. If we import the prog.db into this everything might just crash, I can't imagine how long it would take if I pressed "rebuild all threads" with over 20K threads in the folder.
I could probably write something in PHP in a day that would be equivalent to Shiitchan without being completely shit, but I really hate PHP at this point.
Also, I am >>45 as well. I'm just not going to post with my cap code unless it's an announcement now. I'd say my writing style is recognizable enough, you should be able to figure it out.
I'll be working on the Scheme and FastCGI thing in the next few days to see if I can get a rudimentary textboard up in it. I've got a bit more work to do on the interpreter itself but I guess I'll just host it along with this one and I'll see what bugs are there and fix them.
That's how its called. Unless you are looking at something else.
Technically a simplified web forum
Name:
Anonymous2013-09-04 5:09
I kept feeling like something wasn't right, but couldn't place it. Then it hit me: there are only 10 threads on the front page instead of the standard 18. Please extend.
Name:
Anonymous2013-09-04 5:18
>>45 Do you hate PHP because of PHP4 and Wordpress? Just try a framework like Laravel and see if it's really so bad.
Name:
Anonymous2013-09-04 5:19
Please increase the padding of posts. They look really squashed.
>>50 I noticed that as well. I've remedied it now.
>>52 This isn't something I can fix without rewriting the whole HTML and CSS. The BBS renders differently on Webkit vs Gecko, so it might get fixed when I decide to switch to my own BBS.
>>51 I hate PHP because I hate PHP. Its implementation, the entire language, everything. A framework on top of a broken language isn't going to fix anything. The only plus the language has is convenience, but it's not an important factor here as I can actually program. As I've stated before, I'm working on finishing my Scheme BBS and we'll test it out soon.
Name:
Anonymous2013-09-04 10:51
A framework on top of a broken language isn't going to fix anything
You'd be surprised
Name:
Anonymous2013-09-04 14:05
>>54 Yeah, you'd be surprised at how it doesn't solve anything and introduces even more bugs and quirks.
Name:
Anonymous2013-09-04 14:56
>>52 No, no. I like it how everything is much more compact now. Please don't.
Name:
Anonymous2013-09-04 16:30
c'mon you could do your own userscripts for things like that!
This isn't something I can fix without rewriting the whole HTML and CSS.
You kidding me? It's all CSS only.
>>57 You don't even need userscripts, userstyles is enough.
Name:
Anonymous2013-09-04 22:26
BBCode tags won't work inside quotes.
> This is a {b]test[/b} shows up as malformed markup and it won't even let me post.
(I can't even post the actual tags because even when using {#}, I get the same error. I think it has to do with Tablecat not allowing you to post any BBCode failures. If I recall correctly, that feature is optional.)
Oh wait, it shows up fine in the Preview button, but it won't show up when I post it?
Well, I can wait for that switch, even if it takes another two decades. Shiitchan was worse in many ways, I'm fine with not using BBCode inside quotes.
---- MSN: xX420LiSPSniPeRXx@msn.com IF YOU WERE KILLED TOMORROW IN A TABLE RELATED ACCIDENT, I WOULDNT GO TO YOUR FUNERAL BECAUSE ID BE SITTING ON A TABLE WE TRUE TABLECATS WE SIT TOGETHER WE SIT TOGETHER ON A TABLE
>>68 Like this: ＿_ﾉﾝ ,ゝ,/^´⌒ヽ ／￣￣￣￣￣￣￣￣￣￣￣￣￣￣￣￣ ﾉ~ ｷ（ﾊﾉ )ﾉ）) | ~~ﾚ从^ヮ^ﾉ,) ＜ AFRICAN AMERICAN く]甘ｲつ | . ムlヽゝ ＼＿＿＿＿＿＿＿＿＿＿＿＿＿＿＿＿ し'ﾉ ---- MSN: xX420LiSPSniPeRXx@msn.com IF YOU WERE KILLED TOMORROW IN A TABLE RELATED ACCIDENT, I WOULDNT GO TO YOUR FUNERAL BECAUSE ID BE SITTING ON A TABLE WE TRUE TABLECATS WE SIT TOGETHER WE SIT TOGETHER ON A TABLE
We should have a satire flag (that only the moderator can see), but overuse of the satire flag will result in a perma-ban. (I'm thinking like, a random number between 1-2<<32, decided at post time.
>>104 Yes, that's stupid. We don't need a new board for that; there aren't nearly enough `suggestions' that could reasonably be lodged and it will just turn into complain nonsense like /q/ or be abandoned like /img/. This thread is fine.
>>106 I have no reason to right now. I think giving users the option of doing so is a good idea. >>108 Personally I have no problems with standalone meta threads, but I think all suggestions should be treated as a continuation of one topic, and therefore should be confined to a single thread (or a single series of threads).
>>109 I'd rather just talk to people on here. I might just make a "contact me" page where you post a comment that only I can see in the backend. I won't reply, though. If anyone talks to me not on the board they won't know that I own the site. That's the way I'd like to keep it. After everything is set up and optimal I'll probably also stop posting as Admin completely or alluding to the fact that I am the Admin anonymously.
>>108 I didn't realize there were that many meta threads. The site has just started, but if this keeps up I'll make a /meta/ and move meta threads to it.
Name:
Anonymous2013-09-05 17:15
Hi, this is not about site problem, but about problem related to your source code package.
I'd like to have a fix for this while retaining the compactness in browsers with graphical interfaces. The previous /pro/g// wasted too much vertical space.
Name:
Anonymous2013-09-06 18:49
>>117 I use the finder '/' to review and select them. I would suggest you make you own CSS, and post here if you do not like Tablecats default one. Everything is in the: <div class="shell" id="threadlist"><div><span class="thread"> http://bbs.progrider.org/prog/style.css
Name:
Anonymous2013-09-06 19:35
>>114 Do you think I care? No, I don't give a fuck. Go wherever you want.
>>115 Yes, I know what a self-signed cert is, please don't dump links for me. I am not going to buy a cert or get a free one, the entire CA model is completely retarded.
I'll make a self-signed cert later, whitelist it, and that's it. I'll figure out where to post the fingerprint later.
>>117,118 Does w3m actually read style sheets? I know links and the like don't. I'll try to fix this, but I can't promise anything since the HTML itself is generated in Perl.
>>131 That's really strange... I haven't modified anything at all as far as the HTML goes.
It's probably something Tabletcat put in as a simple automated (crawler bot) spam prevention.
The first field is named "comment", but the one that actually takes the comment is named "other". "comment" is hidden by CSS rules, it seems, and it shows in links since links doesn't read CSS files.
>>134 Always view the source of your pages. Never allow anything to be downloaded from another server, and never run scripts, unless you can read them (including understand). The default behavior of most browsers ignore XSS, something that is still a serious issue today.
>>135 Just because he has Firefox in the background doesn't mean he's a spammer. Sometimes I use links even if I've got FF open to browse certain sites because it's quicker / more convenient.
Name:
Anonymous2013-09-08 21:25
>>137 I do the same, all the time. Why I am telling you to be careful. I would have assumed already a fault in CSS or functionality on my browser before posting, then viewed the source to confirm my suspicions, since I am using a textbrowser. If >>131 work with textbrowsers, he should have known it by now.
>>135 You mean Chrome. >>134 Yes, the fact that NoScript/Scriptsafe/etc. are only available as browser add-ons is, in my opinion, extremely disconcerting. It places far too much trust in web site owners/developers. And while I understand the sentiment, avoiding all scripts that you can't read or understand is incredibly strict and impractical on today's internet. Whenever I can I will simply avoid sites who use shitty tactics like hiding content behind scripts hosted on advertising servers, but whitelisting scripts from googleusercontent.com or amazonaws.com is necessary to access many otherwise-benign sites. >>137 Absolutely. There are many use scenarios for text-only browsers (I'm using Lynx, btw).
At least you understand why I hate the web SO MUCH, and how evilincompetent companies have become. AH GOD, DO NOT REMIND ME OF THOSE JS EVANGELISTS WITH THEIR PREACHING OF THE NEW WEB 5000.0. Googlamozone are an evil company, and you should ignore them.
>>142 I see. Just looked it up on the wiki. I had not heard about this at all.
Well, I don't work as a web developer or anything even remotely related to websites so I don't care about any of this nonsense, really. The browser I use hasn't been updated in like four years.
>>143 Hehehe. I worked as both, a sysadmin and webdev. I hate it, I hate it. No one even heard of USENET or what they really need, gopher; no one. Sigh. When will companies stop investing in worserer is bestest. I hope soon, before people loose all confidence in everything. Sigh. At the least this song is calming me down: https://youtu.be/BDhg1snT8Ic?t=52m25s It's all turning shit. Serious shit. What was that statistic, that only 4 companies in the USA own all the US cables.
>>144 I've also worked as both a sysadmin and a web developer, but it's been a long time since I've done either of those. These days I get to use C. My sysadmin knowledge stays with me purely because I own a few different servers / VPS' with different OS and that I use Linux as my main OS, as well as sort of needing it here and there when I need to set stuff up for projects, either personal or for work.
Things probably won't get better. I don't care too much, though. I'm mostly isolated from the developments in the software world as a whole. If worse comes to worse we can always just write our own OS to run on shitty hacked-together hardware. It's not like a terminal, compiler, text editor, and cli browser take that many resources to run properly.
>>140,141 I used to use Firefox, but about six months ago it started making my system hang and I couldn't figure out why or how to fix it. I decided it was time to give Chrome a shot. Haven't had any major issues so far. As far as being spyware, everything that phones home to Google can be turned off. Besides, Google can do enough passive tracking on enough web users that A. Even if their knowledge of any particular user is otherwise nil it only takes a few discrete data points to draw a relatively accurate picture of that user, and B. they give zero shits that their web browser data-collection tool has an off button, as the 0.1% of users that turn it off are all pretty much the same, anyway.
I just wish there were more options that just FF, Chrome/Chromium, and Opera. None of them are particularly good.
>>146 There aren't any more options because building a browser is a clusterfuck that no sane person would put themselves through. Pretty much all browsers either use WebKit or Gecko nowdays. The rendering / layout engine is basically the entire browser.
Unless someone comes up with a new one I don't really see much variety popping up.
But at the least open schematics hardware is finally surfacing, with optional integrated wifi capable of IEEE 802.11s and IEEE 802.15. The blimp LTE & WiMax idea is wonderful. I hope it gets implemented, even though there is huge lobbying against it, and legal reprocessing for (fly zone height and all).
The song I fetched while looking for the Touhou MIDI songs, then found a playlist, and that was one of the songs. Still, not one did I enjoy enough to keep (extremely picky). The only one I am debating to keep is this one: https://youtu.be/gWn61o_LVOQ OP's avatar is priceless.
>>151 I once thought about implementing a bbs/forum/whatever where everybody was hellbanned in such a way that every day your IP/username/whatever got assigned to a random pool, and you could only see the posts made within that pool, with some exceptions. It's been part of my quest to implement true forgetfulness for the internet, because the idea is that it would inure users to the idea that information on the site is transient, and some of it would come and go.
>>157 Well, the idea was that it would be at the thread level, not the individual post level. And I wanted to have some sort of extra preservation for if you actually replied to a thread - it would stay around... longer? More frequently?
But yeah, it would be a pain in the neck, and if you really cared you could either run a scraper or join in a merge cluster that would trivially defeat it.
>>158 Sort of reminds me of the anonymous IRC concept where you only receive messages if you actually speak on the network, and you can't see who is there or not, it always looks like one user is in the channel even if multiple nicks speak.
>>145,146 Btw, I am not >>147. There far better browser than using spyware (they even copy your buffer, IoW, they have you passwords/credentials): Dooble GNU IceCat xombrero (my picks) Uzbl (my picks) surf (my picks) (with tabbed, it is heaven) dwb (my picks) jumanji (my picks) Luakit (not my pick) NetSurf Web/Epiphany Galeon Dillo Amaya (for authoring) Swiftweasel Conkeror Arora Midori rekonq Kazehakase
And those are GUI. Nothing beats the regular TUIs: w3m, elinks, links2, Links, Linkx, lynx, cURL, ed browser, wget, etc..
>>158 Too difficult. Esp. since we archive our shit. If you been in the 80's you know nothing lasted on on the internet, and always backed up everything.
