i just looked at the forums code... why so much whitespace?

[TuffGuy]

if there's weren't so many blank lines, the forums would load SOOOOOO much faster. i just saved my thread for future reference, and it's 330kb! :Q:Q:Q

if they removed the blank lines, they could probably cut the file size in half, if not more. imagine what that would do for load times and access time!

(btw, i have a feeling that this thread is gonna be either locked or erased...)

 

[Geekbabe]

Don't you think your concerns and ypur suggestions for fixing things would be better posted over in forum issues or pm'd to Zuni directly ?

 

[Hayabusa Rider]

If locked, only because it is a forum issue, but I dont think anyone will take it personally

 

[yakko]

Why don't you try modifying the code to work without the blank lines and see how it works? I bet they are there for a reason.

 

[XZeroII]

HEY! Maybe if I took all the comments out of my C++ program, it would run faster! Great idea!

 

[TuffGuy]

<< Why don't you try modifying the code to work without the blank lines and see how it works? I bet they are there for a reason. >>


trust me, A LOT of the blanks are pointless. try looking at some UBB or phpBB code and you'll see what i mean.

 

[TuffGuy]

<< HEY! Maybe if I took all the comments out of my C++ program, it would run faster! Great idea! >>


there's NO COMMENTS. just BLANK LINES. why don't you try looking at the code before making a dumb remark like that?

 

[gopunk]

who cares? it loads fast for people who subscribe

 

[TuffGuy]

<< who cares? it loads fast for people who subscribe >>


1) ALL users would benefit from
2) it would alleviate a lot of the stress on the server

can a subscriber compare the file sizes of the subsriber pages and the regular pages? is the coding different between the two, or are the only differences the faster server and the lack of ads?

 

[ThisIsMatt]

This has been brought up before... I can't remember the outcome...or wait...it didn't change, so I guess the outcome was NO.

 

[TuffGuy]

It would be kind of interesting if regular users get a different (bloated) version of the pages, while the subscribers would get a different (unbloated) version of the pages. Basically, it would insinuate that they are trying to INTENTIONALLY "force" people to subscribe. Would you still find that acceptable?

In the link that I provided above, the web page ALONE would take ~1min to load on a 56k modem. And that doesn't even include the images.

 

[joohang]

A lot of Cold Fusion apps have that blank space sh*t.

I personally find this forum an excellent app, but there could definitely be some room for improvement. I not just not sure how to comment because I know nothing behind the architecture of this app.

I once suggested data caching because I experienced some problems with posting - which appeared to be an issue with too much load in the database servers. It'd be nice if they implemented some sort of message queuing capabilities also.

It would also be nice if mods didn't just lock my thread before I even had a chance to tell them why I made such a thread. (It was in forum issues also). If they don't want to hear my opinions on how to improve the forums, I have no idea why such forum even exists. After all, I thought that this was a place for beta test of FuseTalk and Zuni welcomed feedback.

They just said that the problem was bandwidth and locked the thread. So I am just assuming that bandwidth is the only major concern with this app. Getting rid of those damn blank spaces could definitely help.

But then, it works. And subscription is cheap so I don't complain.

 

[Hossenfeffer]

<< It would be kind of interesting if regular users get a different (bloated) version of the pages, while the subscribers would get a different (unbloated) version of the pages. Basically, it would insinuate that they are trying to INTENTIONALLY "force" people to subscribe. Would you still find that acceptable? >>

One: they could do whatever they want.
Two: they're not dishing out a bloated version of the page(s). The bandwidth is capped on the non-subscriber servers.
Three: Remember the first two.



<< I thought that this was a place for beta test of FuseTalk . . . >>

Key word there... was. The beta period is over. You can send suggestions to the actual fusetalk site itself.

 

[Gaard]

<< Discuss any issues you are having with the new forums here. You are not guaranteed a response. >>

 

[Hossenfeffer]

<< if there's weren't so many blank lines, the forums would load SOOOOOO much faster. i just saved my thread for future reference, and it's 330kb! >>

Translation that's more apt to get some response (if any): " I just saved my thread for a future reference and it appears as though it takes up 330kb on my hard drive. This seems like a lot and got me thinking about things. If all these 330k (or however big) pages are being served up to people, it would probably be a huge bottleneck. I'm interested in seeing things move as quickly and efficiently as possible and thought it worth my time to ask about the blank lines included in the code. Do the blank lines slow things down and could the code be trimmed to better streamline the use of bandwidth?"

Keyword here: appears.

I could be off the mark, but it's my understanding that the pages are compressed before they're sent out. This way, the blank lines add little, if any, to the overall size of the page(s) being served. Compress a word document with winzip and see how small you can get it. Add a bunch of blank lines and try it again.

 

[gopunk]

first off, no, subscribers do not get any less bloated pages.

secondly, do a little more research.

ZA will tell you how much data is transferred.

http://forums.anandtech.com/messageview.cfm?catid=38&threadid=767924&STARTPAGE=1

that is our test page.

right before i click on the page, i am at 15.53 KB. after i click and load the first page (mine is set to 100 posts), i am at 57.81 KB. when i save the page, the page alone (just code, no images) is 404 KB. this should mean something to you. i mean use your head! i don't know about you, but the notion that for every three pages we see we have to transfer 1MB seems kind of recockulous.

 

[FelixDeCat]

>recockulous

 

[Jason Clark]

Ok, time for a coldfusion lesson, and HTTP compression lesson

ColdFusion server is what puts the blank spaces into the code, not the code, not the product. When coldfusion processes the raw code it replaces most of it with blank space, as it interprets it. TuffGuy, the blank lines would equal a carridge return (Carridge Return CR = 1 byte, Line Feed LF = 1 byte), so even if you spent time and removed them all you wouldn't save as much as you might think. You can use some space limiting tags that are built into coldfusion to try and stop it, but they don't work all that well, and for the most part make a royal mess of code. Sorry, but for anyone that has programmed for awhile, turning your code into a mess over a few KB is not very bright. Not only for the people that write the code, but for the customers that buy source code to modify it.

The second lesson is HTTP compression, TuffGuy, unless you are not using a HTTP 1.1 compliant browser you are not getting 330kb of data. You are getting probably 10-20k. We use HTTP Compression here as most sites should!. All html content is gzipped (built into every http 1.1 browser) and then sent to the client, your browser unzips it. TuffGuy, when you see all the spaces in the browser, those are NOT just wasted blank lines, that is interpreted code. You cannot see source code by just doing view source, you would have to have a source copy of the product to see just what is behind the code. ColdFusion is an interpreted language, in that the source code is "compiled" into pcode and then executed and handed off to the webserver. View source does nothing more than show you the generated code.

So if you have an issue with coldfusion and whitespace, take it up with Macromedia . I agree that it would be nice if Macromedia would do something about it, but we can't all have what we want.

Hopefully that clears some stuff up!.

Wanna read more:

http://webreference.com/internet/software/servers/http/compression/

http://www.w3.org/Protocols/rfc2616/rfc2616.txt

 

[Hossenfeffer]

Zuni, thanks for the lesson. (And thanks to the mod for the new thread title.)

 

[ThisIsMatt]

spanked.

 

[Jason Clark]

Not at all, just wanted to really explain how it all works. . It's sometimes too easy to give quick "Just Cause" answer, I'm guilty of that as are most people at some point. This, hopefully explains it clearly.

Cya!

 

[gopunk]

thanks for the explanation, zuni.

 

[TuffGuy]

thanks for the reply Zuni.

but why is off-topic.net still fast and free while AT is experiencing such problems?

 

[Hossenfeffer]

Why? for one:
@10:20 PST on Off-topic.net:

<< There are currently 247 members and 34 guests on the boards. | Most users ever online was 817 04-23-2002 at 05:35 AM. >>

@Anandtech forums:

<< There are currently 443 users logged in >>



Granted, Saturday night 10 o'clock PST is far from "prime time", but I wouldn't be surprised if you saw a similar trend. The Anandtech figure doesn't take into account the number of lurkers, whereas the off-topic.net figure does.

 

[Scootin159]

Is there a way to tell if a site is zipped when it is sent or not? I basically want to check if my websites are zipped when they are sent....but since it's on a hosting thing I can't really go into the server & check, so I need some kind of browser-side way to test.