|
teh website |
|
This is a long thread. Click here to view the threaded list. |
|
Tony Haines |
Message #119031, posted by Loris at 17:56, 9/10/2011 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
So I know this will probably be old-hat to you guys, but I've been trying to polish my website (which is on shared hosting). I'd appreciate it if you'd have a look and let me know if you think I've missed anything obvious.
http://lysisgames.com/
1) Learned about .htaccess ; Now have 404 and 500 error pages, and hidden directory listings
2) Redone the games pages (all except Big Fish, anyway). Previously they used tables, now I'm using css, (and a shedload of divs). These pages use SSI, which will hopefully make maintainance and adding new ones easier. If you block ads you might wonder at the blank space - Google now request that game pages have 150 pixels separating ads from the game; that's how I arrived at the layout. I considered having a seperate stylesheet, but decided there wasn't enough data there to be worth it. Oh, the link units change colour 'randomly', which is a hack; the ability to natively change ad colours was removed so I've used my newfound SSI functionality to do it anyway. Like I say, probably old-hat, but I'm trying to make progress.
So what needs fixing, what have I missed, or what should I do next? |
|
[ Log in to reply ] |
|
Patric Aristide |
Message #119032, posted by Tin Hat at 18:52, 9/10/2011, in reply to message #119031 |
Member
Posts: 56
|
Is it supposed to look like this:
http://img695.imageshack.us/img695/129/screengraby.jpg |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119033, posted by filecore at 19:19, 9/10/2011, in reply to message #119032 |
Posts: 3867
|
Is it supposed to look like this:
http://img695.imageshack.us/img695/129/screengraby.jpg Looks pretty much the same on Win7/FF3.6 and IE9, except all the boxes are the same size. Is this intentional design? |
|
[ Log in to reply ] |
|
Andrew Rawnsley |
Message #119034, posted by arawnsley at 20:35, 9/10/2011, in reply to message #119033 |
R-Comp chap
Posts: 597
|
The gaps seem to have google adverts in them, which presumably adblocker scripts are hiding in the screenshot. |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119036, posted by Loris at 22:38, 9/10/2011, in reply to message #119033 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
Is it supposed to look like this:
http://img695.imageshack.us/img695/129/screengraby.jpg Looks pretty much the same on Win7/FF3.6 and IE9, except all the boxes are the same size. Is this intentional design? Um, yeah. except for the missing ads Andrew mentions (I knew you wouldn't see them, Jason). The boxes are intended to be the same height though, I suppose Oprah considers CSS height specification to be more guidance than rule. Maybe I can fix that with some br/nbsp hacks - that reminds me about another issue with those to do with font height, I'll need to look into both.
Oh, and it looks like you've got Flash inactivated; I guess you wouldn't be playing the games anyway. ![smile](/images/smileys/smile.gif)
But beyond the home-page, have I neglected anything about websites as a whole?
Regarding the game pages, I ran one of them ( http://lysisgames.com/games/distortion-factor.html ) past browsershots ( http://browsershots.org/ ). In most cases it looked fine, for some it fails abjectly - which may be an issue with the browsershots system in some cases, can't really tell. I do know, though, that MSIE6.0 chokes on the pretty much every page of the website, but I don't know why.
[Edited by Loris at 22:40, 9/10/2011] |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119048, posted by filecore at 10:05, 10/10/2011, in reply to message #119036 |
Posts: 3867
|
I do know, though, that MSIE6.0 chokes on the pretty much every page of the website, but I don't know why. Probably because it's a buggy, insecure, and generally pretty crap decade-old browser, which nobody in their right mind should be using, ten years and three major versions later. |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119056, posted by Loris at 15:03, 10/10/2011, in reply to message #119048 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
Probably because it's a buggy, insecure, and generally pretty crap decade-old browser, which nobody in their right mind should be using, ten years and three major versions later. Well, yeah. But why my baulk at site specifically? Many people seem to still be using it, and they'd probably move on if many sites crashed it.
What really rankles is that MS blame my page - claiming it's got errors even though it validates. |
|
[ Log in to reply ] |
|
Paul Vernon |
Message #119065, posted by PaulV at 23:28, 10/10/2011, in reply to message #119056 |
Member
Posts: 135
|
The CSS doesn't validate...
http://jigsaw.w3.org/css-validator/validator?uri=http%3A%2F%2Flysisgames.com&profile=css21&usermedium=all&warning=1&vextwarning=&lang=en
BUT (and it's a big but), as a professional web developer, IME IE6 does not stick to standards at all and the fact that the CSS isn't valid probably has no bearing on the site breaking although I have to confess, I haven't checked too thoroughly this evening.
Not even standards that were released prior to IE6 were implemented correctly by MS. It's box model implementation was truly awful interpreting the rules as set out by the standards incorrectly in many places and weaving in proprietary tech to support MS extensions designed for Intranets etc. causing major unpredictability.
I build sites for large UK corporates where they have thousands of desktop machines in their business. I'm counting the days until they stop using IE6 but even now, it's still the official locked down browser for the majority of the corporates I deal with and it is truly the bane of the web developers life.
I once worked with someone that described working with IE 6 as follows:
It's like building a house and everything looks great until you put the ridge tiles on. At which point, half of the 1st floor ends up on the ground floor and the toilet becomes an outhouse and there's no valid reason why that's the case!
Sorry this doesn't help much but at least you know you're not alone in having problems with IE6 being pants.
Paul |
|
[ Log in to reply ] |
|
Peter Howkins |
Message #119066, posted by flibble at 23:48, 10/10/2011, in reply to message #119065 |
Posts: 891
|
just use tables for layout ... |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119067, posted by filecore at 07:52, 11/10/2011, in reply to message #119066 |
Posts: 3867
|
just use tables for layout ... Seconded.
<table> <tr> <td>Column 1 Box 1</td><td>Column 2 Box 1</td><td>Column 3 Box 1</td> </tr> <tr> <td>Column 1 Box 2</td><td>Column 2 Box 2</td><td>Column 3 Box 2</td> </tr> </table>
etc. Easy, and pretty much guaranteed to work in every browser ever. |
|
[ Log in to reply ] |
|
Andrew Poole |
Message #119071, posted by andypoole at 09:50, 11/10/2011, in reply to message #119067 |
Posts: 5558
|
just use tables for layout ... Seconded.
<table> <tr> <td>Column 1 Box 1</td><td>Column 2 Box 1</td><td>Column 3 Box 1</td> </tr> <tr> <td>Column 1 Box 2</td><td>Column 2 Box 2</td><td>Column 3 Box 2</td> </tr> </table>
etc. Easy, and pretty much guaranteed to work in every browser ever. How very Web 1.0 ![smile](/images/smileys/smile.gif)
On a related note, perhaps you need one of these mugs. |
|
[ Log in to reply ] |
|
Paul Vernon |
Message #119073, posted by PaulV at 10:41, 11/10/2011, in reply to message #119066 |
Member
Posts: 135
|
just use tables for layout ... Sacrilege!!! I've not used tables for layout in years!
Paul |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119074, posted by filecore at 11:07, 11/10/2011, in reply to message #119073 |
Posts: 3867
|
just use tables for layout ...
Sacrilege!!! I've not used tables for layout in years! On a related note, perhaps you need one of these mugs. To be fair: CSS for layout, but tables for tabular data. Being all modern and whatnot does not mean you can no longer use tables! They are not a horrible design hack the way that frames were. And they're standards-complient, cross-browser, yada yada etc. So, hardly sacrilege*.
* yes, many of my websites do use divs for table-type layout rather than using tables. Some of them just use tables. Both types of site validate properly, render properly, and look as intended to the end user. This is all done transparently and with no server or load-time penality - for either variant. Which, really, is what it's all about, isn't it?
Sheesh, I hate purists. |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119076, posted by Loris at 12:08, 11/10/2011, in reply to message #119065 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
The CSS doesn't validate...
http://jigsaw.w3.org/css-validator/validator?uri=http%3A%2F%2Flysisgames.com&profile=css21&usermedium=all&warning=1&vextwarning=&lang=en
Hmmm. I'd not realised/forgotten that the CSS within a page had to be validated seperately. However, it's all valid anyway except this:
width:Expression(document.body.clientWidth > 800? "800px": "auto" );
While it's true that isn't valid CSS, that's there because IE doesn't respect the max-width command which precedes it. I don't think I made any silly errors with it.
I once worked with someone that described working with IE 6 as follows:
It's like building a house and everything looks great until you put the ridge tiles on. At which point, half of the 1st floor ends up on the ground floor and the toilet becomes an outhouse and there's no valid reason why that's the case!
Heh, that's absolutely it. Also that you fudge it with some 'extra tile hack' and it's fine until your grandmother visits, when the rooms become very thin and the furniture reaches up to the ceiling - but only for her.
Can't even remember why I switched from tables now. Except that it's fun to resize the page and watch the layout match the width nicely, which I don't think is possible with tables. Well, maybe it's possible - if I recall correctly I did consider javascript or something might do it - but not sensibly, anyway. Tables would be fine except that apparently you can't rely on combinations in specification (such as making one column scale and another fixed-width, or whatever).
Oh, yeah. I asked people what they thought of the site (a few years ago, elsewhere), and they thought it was too plain. Without exception - so I had to do something. |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119079, posted by filecore at 13:47, 11/10/2011, in reply to message #119076 |
Posts: 3867
|
The CSS doesn't validate... The first thing I did was run that site through W3C, but those errors in the CSS minor niggles are not relevant to the problems you're having, as far as I can tell.
it's fun to resize the page and watch the layout match the width nicely, which I don't think is possible with tables. Well, maybe it's possible - if I recall correctly I did consider javascript or something might do it - but not sensibly, anyway. Uh, <table width="80%"> or similar? You can specify absolute and relative widths for both tables and individual table cells, iirc, and you can also merge sets of cells if you want one which is wider or deeper than the others around it. Not difficult at all.
Not stop being so lazy. |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119080, posted by Loris at 15:07, 11/10/2011, in reply to message #119079 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
Uh, <table width="80%"> or similar? You can specify absolute and relative widths for both tables and individual table cells, iirc, and you can also merge sets of cells if you want one which is wider or deeper than the others around it. Not difficult at all.
Not stop being so lazy. There's no need to be like that. I had a table based layout originally (you can follow the link at the bottom of the index page to a static version I'm still maintaining, which uses one). Yes, tables are easier, but I was trying to get away from them for layout, since that's deprecated - among other reasons. I've probably got several tables in my articles using tables with merged cells, etc. It's not that I don't know how to use tables. Have you tried narrowing your browser screen? See the boxes flow appropriately? I was really pleased with that, and it's not something you can do with tables (without rewriting the table somehow - did I mention I considered that?).
Incidentally: "The width attribute of <td> is deprecated, and is not supported in HTML 4.01 Strict / XHTML 1.0 Strict DTD." http://www.w3schools.com/tags/att_td_width.asp
I originally posted because (among other things) I'd moved my games pages (for example http://lysisgames.com/games/silverlining.html ) away from the previous, horrible, table format to a more flexible css-based one (with SSI making new pages easier.) I can't just use CSS cell widths, (or leave them to auto) because I've got very specific constraints imposed by Google. I guess I could have kept a table for layout, and stuffed it full of shims to force the required minimum widths and heights.
Now obviously, it's frustrating that every browser is broken in it's own special way (CSS: simple, elegant and fucking broken) - but it's not unique to me, and I don't see it as my fault. I do want to know of any issues so I can hack round them, like everyone else does. |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119082, posted by filecore at 21:15, 11/10/2011, in reply to message #119080 |
Posts: 3867
|
I can't just use CSS cell widths, (or leave them to auto) because I've got very specific constraints imposed by Google. Have you thought about simplifying your life by, y'know, not plastering Google adverts all over the site? Since most of world+dog uses adblockers now anyway... I mean, maybe I'm missing something, but how much real-world income does it actually bring you (you specifically)? What's your rationale for implementing it, especially when it's apparently causing such headaches? Genuine interest now. |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119083, posted by filecore at 21:37, 11/10/2011, in reply to message #119080 |
Posts: 3867
|
Incidentally: "The width attribute of <td> is deprecated, and is not supported in HTML 4.01 Strict / XHTML 1.0 Strict DTD." http://www.w3schools.com/tags/att_td_width.asp Umm. What's your point, exactly? You're making your own problems. Use HTML 4 Transitional rather than Strict. HTML 4 Strict deprecated pretty much every single attribute that you could recreate in CSS, so of course it isn't "supported". However, not supported doesn't mean won't work - all the previous HTML is deprecated, and yet it actually works in all browsers. Do you want something that is "standards compliant" by the W3C checker, or something which, on a practical level, actually renders properly for your site visitors?
Incidentally:
<td style="width: 80%;">
Edit: oops, forgot my punctuation!
[Edited by filecore at 22:04, 11/10/2011] |
|
[ Log in to reply ] |
|
Andrew Poole |
Message #119084, posted by andypoole at 21:38, 11/10/2011, in reply to message #119082 |
Posts: 5558
|
I can't just use CSS cell widths, (or leave them to auto) because I've got very specific constraints imposed by Google. Have you thought about simplifying your life by, y'know, not plastering Google adverts all over the site? Since most of world+dog uses adblockers now anyway... I mean, maybe I'm missing something, but how much real-world income does it actually bring you (you specifically)? What's your rationale for implementing it, especially when it's apparently causing such headaches? Genuine interest now. ^ This. IME, unless you've got a fairly decent amount of traffic coming to your site, Google Ads won't earn enough to be useful. I had some Google Ads on a small site for over a year and made about £1. Google don't pay up until you get to £50 (IIRC).
Plus, as Jason said, most of the world has ad blocking software these days (It's generally the first thing I install in my browser), so will just see empty wasted space. |
|
[ Log in to reply ] |
|
Andrew Poole |
Message #119085, posted by andypoole at 21:40, 11/10/2011, in reply to message #119083 |
Posts: 5558
|
Incidentally:
<td style="width: 80%"> Yep. <td width="80%"> and <td style="width: 80%;"> are, to all intents and purposes, identical in function.
Although Jason forgot his semicolon ![wink](/images/smileys/wink.gif) |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119091, posted by Loris at 13:08, 12/10/2011, in reply to message #119083 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
Do you want something that is "standards compliant" by the W3C checker, or something which, on a practical level, actually renders properly for your site visitors? Both, actually. The first ought to imply the second.
I know I'm skipping a few of your points, but then you seem to have missed most of mine, so we're even. ![smile](/images/smileys/smile.gif)
The most important of mine was that I can't just use a table, it would need a load of fudging elements to be viable - and even then it may well look bad in some browsers (eg. if the 1x1 transparent image doesn't load for some reason, a browser may decide to fill the space up with a load of crap).
Regarding why the ads, I think I'm not actually allowed to say how much it's taken exactly. I think I'll be alright saying that the minimum payout is £60, and I've not quite reached it. So no, it's nothing to write home about. However, the thing is that if a game is a big hit you can get a lot of traffic and it can do quite well. I've got several games 'finished' which haven't been sponsored, so ads are pretty much my only option for them. |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119092, posted by filecore at 13:46, 12/10/2011, in reply to message #119091 |
Posts: 3867
|
1x1 transparent image I'm lost for words ![big smile](/images/smileys/smile.gif) |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119152, posted by Loris at 12:28, 18/10/2011, in reply to message #119092 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
I'm lost for words ![smile](/images/smileys/smile.gif) I'm not sure what you mean by that, but I would like to know.
I think you're suggesting that such shims are unnecessary, but I can't see a way of doing what I need without them.
For my games pages, I basically need the right column to be fixed width, and the left side to be at least a particular width, expanding to fill the browser screen - ideally up to a maximum width. Now, I can't just define the right side's width and leave the left to auto, because then it may become too thin. (Plus it breaks the rule that if you specify one fixed-width you need to specify them all.) Doing it with shims should be possible, but it's inelegant to say the least - and I suspect that it makes a mess occasionally as I described.
And I think you're implying that there's a better way. So what am I missing? |
|
[ Log in to reply ] |
|
VinceH |
Message #119156, posted by VincceH at 19:32, 18/10/2011, in reply to message #119152 |
![VincceH](/images/avatars/574.png) Lowering the tone since the dawn of time
Posts: 1600
|
And I think you're implying that there's a better way. So what am I missing? http://www.w3schools.com/cssref/pr_dim_min-width.asp |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119161, posted by Loris at 13:24, 19/10/2011, in reply to message #119156 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
IE6 doesn't respect min-width. |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119163, posted by filecore at 14:20, 19/10/2011, in reply to message #119161 |
Posts: 3867
|
IE6 doesn't respect min-width. I think I see your problem. |
|
[ Log in to reply ] |
|
Tony Haines |
Message #119164, posted by Loris at 15:18, 19/10/2011, in reply to message #119056 |
Ha ha, me mine, mwahahahaha
Posts: 1025
|
Well, yeah. But why my baulk at site specifically? Many people seem to still be using it, and they'd probably move on if many sites crashed it.
What really rankles is that MS blame my page - claiming it's got errors even though it validates. Ahah! Possible explanation for IE6 crashes:
http://css.flepstudio.org/en/css-tips/min-width-ie6.html regarding the property expression hack to make IE6 behave: width:Expression(document.body.clientWidth < 502? "500px": "auto" ); ... "NB: to avoid that IE6 crashes using the above method it is important that the first value (502 in this example) is different from the second one (500 in this example) and that no margin or padding are applied to the div ‘container’."
So it actually is that expression stuffing it - funny (annoying) that my original source for it didn't mention that.
...or alternatively, they suggest a shim; they propose an empty fixed-width div. I can't say I like either fix, but which do/should I dislike least? |
|
[ Log in to reply ] |
|
Martin Bazley |
Message #119169, posted by swirlythingy at 01:27, 20/10/2011, in reply to message #119164 |
Posts: 460
|
Make it break on IE6 and leave the mouth-breathers who use it to rot in the past. |
|
[ Log in to reply ] |
|
John Hoare |
Message #119170, posted by moss at 01:51, 20/10/2011, in reply to message #119169 |
Posts: 9348
|
I haven't supported IE6 - or, at least, correct visual formatting on IE6 - on any of my sites for years now. Never had a single complaint. |
|
[ Log in to reply ] |
|
Jason Togneri |
Message #119171, posted by filecore at 05:37, 20/10/2011, in reply to message #119170 |
Posts: 3867
|
Quite. Indeed.
That said, it's probably fairly typical of RISC OS users and developers to be stuck in the past. I suppose being hung up on buggy crap from "only" a decade ago is probably a step forward. At least it's not buggy crap from two decades ago. |
|
[ Log in to reply ] |
|
Pages (2): 1
> >|
|