Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

"Unknown or expired link." - Why?

google.com

Post reply on HN
Print view View original

131–140 of 142 posts

Re: "Unknown or expired link." - Why?

#131

Post by gnaritas »

dasil003 wrote: post #127
gnaritas wrote: post #114

Earlier quoted context omitted.

You seem to not be aware of the architecture of this site, it's all run out of ram, no database; just simple lazy load on demand files on a single server running in a single process. Because of this, memory is tight, that's why those closures are purged. It simply means pg hasn't had the time to convert more of the prototyped code into production stateless code that always works. But that's his prerogative, this site…

There's something on disk isn't there? Or does a power outage mean poof it's all gone? Anyway, you're totally right that it's his prerogative to build a site however he sees fit, and it's my prerogative to leave, but it's also my prerogative to complain about it and call it half-assed. I do build websites as well, so I'm not just armchair commenting.

> Or does a power outage mean poof it's all gone?

Poof the closures are all gone; the state of the articles and comments are of course rebuilt from disk state on an as needed basis.

> I do build websites as well, so I'm not just armchair commenting.

I appreciate that, I just use a similar framework and understand why one would choose to use callbacks and not bother ever replacing them. It has to matter enough to bother and to pg, it doesn't yet.

Re: "Unknown or expired link." - Why?

#132

Post by wvenable »

gnaritas wrote: post #130
wvenable wrote: post #120

Earlier quoted context omitted.

I certainly don't think 3 hours is enough for this site but then pg has put in a lot more time in than just his initial prototype as well. But the functionality of closures and server-side state aren't needed for this kind of site at all. Nothing you mentioned above seems related to that functionality either.

It's not relevant whether they're needed or not if you use closures as your default linking mechanism because it's always the easier route. They're not needed for any site ever, they're just damn convenient. You go back through the app replacing closures with more verbose and direct linking as time permits; clearly pg hasn't found the time yet.

You could route URLs for this site with a moderately sized switch statement. The code to route everything would be smaller than the code needed to just route closures before you've setup a single one.

Obviously replacing closures with directly linking is more difficult than just using direct linking in the first place.

Re: "Unknown or expired link." - Why?

#133

Post by holdenk »

JoachimSchipper wrote: post #26
piinbinary wrote: post #16

Earlier quoted context omitted.

I would be interested to know how much state is in those closures. If it is less than 200 or so bytes, it would not be impractical to encode it (b64) in the url for the next page (rather than a reference to the state).

You don't want to execute code from URLs. (Yes, you can use cryptography to "sign" URLs you create. Don't try that at home unless you know the difference between MACs and hashes, and how to avoid timing attacks.)

The PLT-Scheme (aka Racket) folks do this. Its kind of awesome, and I used it for a side project. It works pretty well except some browsers have a max url length, so you can only safely shove so much data to the client side, and then your stuck with old fashion IDs again. Still lots of fun to play with :)

In a scary way

Re: "Unknown or expired link." - Why?

#134

Post by mooism2 »

gnaritas wrote: post #17
mooism2 wrote: post #11

The HN server uses a table of closures to implement those links (the id code for the closure is the bit after fnid= in the url). When the HN server starts running out of memory, it drops entries from this table. When your browser asks for an entry that is no longer in this table, you get the "Unknown or expired link" error. This is a crazy design, but unless someone would like to patch the source code and get PG to a…

It's not crazy at all, it greatly simplifies development to use callbacks for actions rather than manually encoding the necessary state into the URL. Techniques such as this are what enable a single developer to be so productive by automating boring and time consuming stuff.

You're justifying code that doesn't work on the grounds that it was quick to write. (Facetious comment: if the website doesn't have to work, I can write the whole thing in under a minute. Someone wrote a HN clone on these lines a few weeks back, but I don't know how long it took them.)

Given that the HN code was written by an increasingly busy man in his spare time, his use of an unreliable but quick-to-write implementation technique may be an acceptable trade-off. But it doesn't make the design any less crazy.

Re: "Unknown or expired link." - Why?

#135

Post by Tichy »

stickfigure wrote: post #80
Tichy wrote: post #20

Earlier quoted context omitted.

Hacker News is open source, so it seems as if nobody is interested in fixing it. Or are there fixes and pg has rejected them?

Hacker News is open source Is this true? If so, where can I find the code?

I think it is included with arc (the language).

Re: "Unknown or expired link." - Why?

#136

Post by gnaritas »

wvenable wrote: post #132
gnaritas wrote: post #130

Earlier quoted context omitted.

It's not relevant whether they're needed or not if you use closures as your default linking mechanism because it's always the easier route. They're not needed for any site ever, they're just damn convenient. You go back through the app replacing closures with more verbose and direct linking as time permits; clearly pg hasn't found the time yet.

You could route URLs for this site with a moderately sized switch statement. The code to route everything would be smaller than the code needed to just route closures before you've setup a single one. Obviously replacing closures with directly linking is more difficult than just using direct linking in the first place.

Incorrect; routing closures is free, automated by the web framework, something no web framework does for direct linking. Look at Rails, every Rails programmer spends a large amount of time figuring out and managing routing. Look at Seaside which uses callbacks just like this site, you can build the entire app without spending a second on routing URL's because callbacks do it for free as part of the framework, completely automated.

Re: "Unknown or expired link." - Why?

#137

Post by gnaritas »

mooism2 wrote: post #134
gnaritas wrote: post #17

Earlier quoted context omitted.

It's not crazy at all, it greatly simplifies development to use callbacks for actions rather than manually encoding the necessary state into the URL. Techniques such as this are what enable a single developer to be so productive by automating boring and time consuming stuff.

You're justifying code that doesn't work on the grounds that it was quick to write. (Facetious comment: if the website doesn't have to work, I can write the whole thing in under a minute. Someone wrote a HN clone on these lines a few weeks back, but I don't know how long it took them.) Given that the HN code was written by an increasingly busy man in his spare time, his use of an unreliable but quick-to-write impleme…

> You're justifying code that doesn't work on the grounds that it was quick to write.

No, the code works fine for an acceptable period of time, it doesn't just not work.

Re: "Unknown or expired link." - Why?

#138

Post by wvenable »

gnaritas wrote: post #136
wvenable wrote: post #132

Earlier quoted context omitted.

You could route URLs for this site with a moderately sized switch statement. The code to route everything would be smaller than the code needed to just route closures before you've setup a single one. Obviously replacing closures with directly linking is more difficult than just using direct linking in the first place.

Incorrect; routing closures is free, automated by the web framework, something no web framework does for direct linking. Look at Rails, every Rails programmer spends a large amount of time figuring out and managing routing. Look at Seaside which uses callbacks just like this site, you can build the entire app without spending a second on routing URL's because callbacks do it for free as part of the framework, complet…

You get crappy URLs with closures; no other web framework provides ugly URLs like that -- but if they did -- they could do it with the same amount of effort. But routing in rails is specifically designed to separate URL presentation from the underlying action.

Hell, you can get routing for free in PHP if you just name your files like: comment.php, topic.php, upvote.php, downvote.php, etc.

Re: "Unknown or expired link." - Why?

#139

Post by gnaritas »

wvenable wrote: post #138
gnaritas wrote: post #136

Earlier quoted context omitted.

Incorrect; routing closures is free, automated by the web framework, something no web framework does for direct linking. Look at Rails, every Rails programmer spends a large amount of time figuring out and managing routing. Look at Seaside which uses callbacks just like this site, you can build the entire app without spending a second on routing URL's because callbacks do it for free as part of the framework, complet…

You get crappy URLs with closures; no other web framework provides ugly URLs like that -- but if they did -- they could do it with the same amount of effort. But routing in rails is specifically designed to separate URL presentation from the underlying action. Hell, you can get routing for free in PHP if you just name your files like: comment.php, topic.php, upvote.php, downvote.php, etc.

Users don't care what the URL looks like, Amazon does quite a bit of business with its crappy URL's and closures allow linking to actions without having to have a resource for every action. This is extraordanaryily useful when building complex applications.

Look, I use both styles daily and I'm telling you it'll be over my dead body before I allow someone to take closure style ugly links away from me. You aren't going to convince me that manually routed URL are always preferrable.

Re: "Unknown or expired link." - Why?

#140

Post by wvenable »

gnaritas wrote: post #139
wvenable wrote: post #138

Earlier quoted context omitted.

You get crappy URLs with closures; no other web framework provides ugly URLs like that -- but if they did -- they could do it with the same amount of effort. But routing in rails is specifically designed to separate URL presentation from the underlying action. Hell, you can get routing for free in PHP if you just name your files like: comment.php, topic.php, upvote.php, downvote.php, etc.

Users don't care what the URL looks like, Amazon does quite a bit of business with its crappy URL's and closures allow linking to actions without having to have a resource for every action. This is extraordanaryily useful when building complex applications. Look, I use both styles daily and I'm telling you it'll be over my dead body before I allow someone to take closure style ugly links away from me. You aren't goin…

Users who bookmark and search engines care about non-random non-expiring urls. The urls on HN are wrong in every way a url can be wrong.

Closures create a resource for every closure instance which isn't very scalable or efficient -- it is in fact the problem with this very site. It might be useful for building complex applications but it's just a liability and a waste for something as simple and busy as HN.

I'm trying to convince you that manually routed URLs are always preferable but they should be for a site like this.

Post reply on HN