An amateur has tried to build a website in raw HTML

A place to discuss the implementation and style of computer programs.

Moderators: phlip, Moderators General, Prelates

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Mon Apr 02, 2018 6:38 pm UTC

Hey, guys!
I've tried to make a website in raw HTML (using only the tools such as Notepad++), but I am not a professional programmer or a web-designer. You can see it here:
http://flatassembler.000webhostapp.com/
(It sleeps every day from 22h to 23h UTC+2.)
Do you have some ideas on how to improve it that are easy to program?
Thanks in advance!

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3502
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: An amateur has tried to build a website in raw HTML

Postby Soupspoon » Mon Apr 02, 2018 7:06 pm UTC

Without going to far into your javascript (which is where you're programming, and I haven't yet strayed beyond the front page) a quick look at the source pleases me in its neatness (doing it by hand, it is lacking a lot of the usual WYSIWYG Web Designer Program awfulness that creeps in there).

First thing I'd suggest, though, is to take your in-line CSS section and chuck it in a .css file of its own. Assuming you're not making every page on your site completely different in style, that one file can serve every one of them. (Truly unique formatting on each can be specified in a shortened in-line CSS section or specified by special id/class overrides within the 'global' file. Your choice.)

That's from a quick glance, only. I'm going back in a moment to work out what that floating button is, and look into some of the linked subpages, and perhaps see if there's anything that does/does not work particularly well with this tablet version of the browser (but I think you've not invoked some of the sillier stuff - I shall know more shortly). But I like your writing style. I'm a notepad/equivalent user, myself, when it comes to writing HTML, and much appreciate it being structured for human-readability.

edit to add: the Croatian variable names (I discover that "krug" is likely "circle") is confusing in the code, but not a problem unless/until you co-develop with someone without any Croatian (and without some Google Translate ability?!? :P). I like the analogue clock, calculator and dragon curve. Haven't spotted any obvious problems with the programming (in-between the weird variable names!) and also quite neat so shouldn't be too hard to understand if I try a bit harder. I shall delve further later, and maybe someone else might have spotted something good/bad that I've missed, by then, or just have a different view from mine.

User avatar
Jplus
Posts: 1711
Joined: Wed Apr 21, 2010 12:29 pm UTC
Location: Netherlands

Re: An amateur has tried to build a website in raw HTML

Postby Jplus » Mon Apr 02, 2018 8:18 pm UTC

I have a suggestion which is not directly "easy to program" because you have to learn some new things first, but: in the long term, you can make programming (in JavaScript) easier for yourself by using some libraries. You could start with jQuery and Lodash.
"There are only two hard problems in computer science: cache coherence, naming things, and off-by-one errors." (Phil Karlton and Leon Bambrick)

coding and xkcd combined

(Julian/Julian's)

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Tue Apr 03, 2018 7:51 am UTC

We were using JQuery in the web-game about linguistics. It makes it easy to add flashy-yet-cheap animations. Not all code there is mine, some code is from Daniel Ross (the linguist who made linguistforum.com) and some code is from Boris Muminovic (a programmer from the Croatian web-development company Osvit).

Any ideas why Internet Explorer 11 and Opera 36 appear to ignore half of the CSS on my homepage? Also, why do the JavaScript-generated SVGs look whitish in, well, all the browsers I've tested my website in except Safari 6.2 (that is the browser I am usually using)?

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Tue Apr 03, 2018 8:31 am UTC

hm, if you want the text on your site to actually be readable, you maybe shouldn't make it light gray (or any other colour really) on a garish background image.

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Tue Apr 03, 2018 10:26 am UTC

speising wrote:hm, if you want the text on your site to actually be readable, you maybe shouldn't make it light gray (or any other colour really) on a garish background image.

Which browser are you using? As far as I know, it's only rendered that way in Internet Explorer. All the other browsers put a dark gray layer in front of the background image and behind the text (at least in the Style #1, there are two other styles that can be chosen in the bottom right corner). I won't spend time making it work in IE (which almost nobody uses today) unless there is a really quick fix.

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Tue Apr 03, 2018 10:39 am UTC

FlatAssembler wrote:
speising wrote:hm, if you want the text on your site to actually be readable, you maybe shouldn't make it light gray (or any other colour really) on a garish background image.

Which browser are you using? As far as I know, it's only rendered that way in Internet Explorer. All the other browsers put a dark gray layer in front of the background image and behind the text (at least in the Style #1, there are two other styles that can be chosen in the bottom right corner). I won't spend time making it work in IE (which almost nobody uses today) unless there is a really quick fix.

hm, why have that background image at all then? you can barely see it in chrome.

Tub
Posts: 383
Joined: Wed Jul 27, 2011 3:13 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby Tub » Tue Apr 03, 2018 1:19 pm UTC

Jplus wrote:I have a suggestion which is not directly "easy to program" because you have to learn some new things first, but: in the long term, you can make programming (in JavaScript) easier for yourself by using some libraries. You could start with jQuery and Lodash.

jQuery is largely obsolete (unless you want to support IE8 or something). Modern browsers have native APIs that do the same, just faster, better and more consistent. Use native selectors instead of $('.myclass'). Use css transitions/animations instead of jquery anims. Use DocumentFragment (or an actual framework with templating) instead of $('<div>text</div>'). IMHO learning jQuery in 2018 is a waste of time.

User avatar
Jplus
Posts: 1711
Joined: Wed Apr 21, 2010 12:29 pm UTC
Location: Netherlands

Re: An amateur has tried to build a website in raw HTML

Postby Jplus » Tue Apr 03, 2018 8:47 pm UTC

Tub wrote:jQuery is largely obsolete (unless you want to support IE8 or something). Modern browsers have native APIs that do the same, just faster, better and more consistent. Use native selectors instead of $('.myclass'). Use css transitions/animations instead of jquery anims. Use DocumentFragment (or an actual framework with templating) instead of $('<div>text</div>'). IMHO learning jQuery in 2018 is a waste of time.

I respectfully disagree. Yes, a lot of the functionality in jQuery is nowadays also available through the web standards, but jQuery still does a good job at shortening syntax and ironing out differences between browsers. (By the way, jQuery uses native selectors when available and documents it when a selector is nonstandard). jQuery in itself is consistent and predictable. And then there is a mountain of plugins for jQuery, many of which are mature and useful. I'd say that jQuery is an industry standard.

I do agree that using proper templating is a good idea and also that using a framework is a good idea. Using a framework or templates, however, does not preclude using a couple of additional toolkits such as jQuery. I was just mentioning jQuery and Lodash because I think they're very useful and they're only one step away from what FlatAssembler already knows.
"There are only two hard problems in computer science: cache coherence, naming things, and off-by-one errors." (Phil Karlton and Leon Bambrick)

coding and xkcd combined

(Julian/Julian's)

Tub
Posts: 383
Joined: Wed Jul 27, 2011 3:13 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby Tub » Wed Apr 04, 2018 12:20 am UTC

Warning, strong opinions about jQuery ahead.
Spoiler:
I did use jQuery a lot back in the day, when it had its uses. But believe it or not, for a couple of years I've been programming without ever thinking "This problem would be easier to solve with jQuery." I haven't had much issues with browser differences, either. IE is trouble, as usual, but the trouble I had would not have been solved by jQuery.

The issue I have boils down to the idea of keeping state in your DOM. You want to add a mouseover-animation to all headlines, but you don't know what headlines you have unless you look into the DOM, so you go $('h1').onMouseOver(...).
If you keep your state in javascript, you render those headlines, and you attach the event listeners at the same time. Keeps related code close. I barely use selectors in javascript, usually just one per webapp to specify the root element, maybe another if I need to interface widgets from different frameworks.
Selectors are a proper bandaid if you're generating half the page on the server and half the page on the client - which is kinda reasonable if all you have is PHP and jQuery, but we have better tools now.

It gets worse with animations. Css transitions have a clean API, they keep your styles in your css (not in your javascript), they're performant, they work concurrently and they're easy to reason about. Same for css animations, if you need something a little more complex. Doing simple animations with jQuery will always lead to code that's more verbose and less readable.

But where jQuery really gets horrible is when you try complex animations. Because, again, it doesn't properly track state, and half your state ends up in the DOM. It gets increasingly difficult to reason about when to call .stop() or .finish() before .animate() and when not. Having two concurrent animations on a single element, even on different attributes, will simply not work the way you want it to, so in the end you give up and ignore any clicks while an animation is running because otherwise your layout will break.
As it turns out, calling requestAnimationFrame and an easing function on your own isn't terribly difficult, and I've found complex and interlocking animations to be much easier to write manually with proper state management than to squeeze them into the jQuery animation API.
Jplus wrote:jQuery in itself is consistent and predictable.

Also, I've had half of my animations break when upgrading to a newer jQuery version. No, there were no documented API changes for that release. But in jQuery's defense, they broke consistently on all browsers.

Those are the three major features - cross browser compatibility, selectors and animations. I have no use for the first, because I don't need to support IE8 any more. I have no use for the second, because I refuse to split html generation code between server and client. And I really don't want to touch the animation API ever again.
Jplus wrote:And then there is a mountain of plugins for jQuery, many of which are mature and useful. I'd say that jQuery is an industry standard.

You don't need to learn jQuery to use a plugin based on jQuery. Also, your basic calendar widget and image slideshow, there's like a million of those, for every single framework, and even without framework dependencies. If the plugin is useful, then it's probably not jQuery exclusive.

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Wed Apr 04, 2018 8:48 am UTC

JQuery is even today the most used JavaScript framework by a large margin. It's nice to be able to understand the old code in the language you use, even if JQuery itself is relatively useless today.

User avatar
Flumble
Yes Man
Posts: 2023
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby Flumble » Wed Apr 04, 2018 5:50 pm UTC

Soupspoon wrote:First thing I'd suggest, though, is to take your in-line CSS section and chuck it in a .css file of its own. Assuming you're not making every page on your site completely different in style, that one file can serve every one of them.

I subscribe to this too. And next to CSS, I also advise you to move the Javascript to their own files.

Speaking of Javascript: the best way (as far as I know) to include your scripts (once you've extracted all the inline javascript and put it into files), is to put <script src="someFile1.js" defer></script> the <head>. Putting it in the <head> has the advantage that the browser will load the scripts while it's busy parsing the rest of the page. Adding defer makes sure your script is run once the browser is done parsing the page (so, for example, gornjiLijevi exists).
Another point: your animations. For some time now, decent browsers are supporting requestAnimationFrame. This usually updates far more frequently than a setInterval, so your animation becomes smoother. You do have to change your gravity constant of course (and possibly use the time difference to get a framerate-independent simulation :D ).


Next, it's nice to see you're using HTML5 tags. To continue this, I suggest you put your navigation links inside a <nav> element (either inside of the <aside> or instead of the <aside>; the internet can't seem to make up its mind about having <nav> inside <aside> or not). And the asides you have on the right side of the pages, fit best inside an <aside> rather than a <section>.

And I suggest you use a tool like WAVE to check if your semantical HTML makes sense (most notably for screen readers –you know, all those people who can't read your linguistic journeys, but can only hear them): http://wave.webaim.org/report#/https://flatassembler.000webhostapp.com/.
For example, it says you're missing a <h1> tag, which you can easily fix because the <div> with "FlatAssembler's Homepage" is actually your level 1 header. This will also make CSS styling of your page title easier. Well, that is, if your header was simply one page-wide element with a scaling SVG background (meaning you'll have to make/generate an SVG file rather than generating it with javascript). In the current state, the "cleanest" way to have your title would be to put a <h1> inside the <div> and have the following CSS:

Code: Select all

header
{
/*header only has 1 rule left, because only the height really is a property of the whole header*/
  height: 200px;
}

/*note the "header h1": your page *may* have a <h1> for things likes <article>s too*/
header h1 {
/*the head text itself gets most of the header's styling*/
  text-align: center;
  color: darkblue;
  font-size: 36px;
  font-weight: bold;
  font-family: Lucida;

/*and flexbox magic to vertically center the text regardless of whether the title gets split into 2 lines or you resize the header to anything other than 200px height*/
/*source: https://css-tricks.com/centering-css-complete-guide/ */
  display: flex;
  justify-content: center;
  flex-direction: column;
  height: 100%;
  margin: 0;
}


WAVE also notes that your homebrew links within the linguistics link are inaccessible for screen readers ...and basically broken for normal users: you don't want to both go to the linguistics page and go to the toponyms page in another window at once.


speising wrote:hm, if you want the text on your site to actually be readable, you maybe shouldn't make it light gray (or any other colour really) on a garish background image.

What browser doesn't show the background: black; behind the main text? :?
I'm inclined to complain about the contrast in the navigation though. Brown on grey and light blue is hard to read.

User avatar
Xanthir
My HERO!!!
Posts: 5311
Joined: Tue Feb 20, 2007 12:49 am UTC
Location: The Googleplex
Contact:

Re: An amateur has tried to build a website in raw HTML

Postby Xanthir » Fri Apr 06, 2018 2:49 am UTC

Flumble wrote:
Putting it in the <head> has the advantage that the browser will load the scripts while it's busy parsing the rest of the page.

This is not true - the location of your script has no effect on how it interacts with rest-of-page parsing. In fact, it was common advice to put your JS at the *bottom* of your page for a long time, because by default scripts pause the parser while they're downloading and executing, and putting it at the end meant your page at least got parsed and displayed before the script started interrupting things.

These days you have the "async" and "defer" attributes to help with that; see the SO answer (with a useful diagram!) for details on what each does, so you can decide which one you need. (Note that async overrides defer, if you specify them both.)
(defun fibs (n &optional (a 1) (b 1)) (take n (unfold '+ a b)))

User avatar
Flumble
Yes Man
Posts: 2023
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby Flumble » Fri Apr 06, 2018 10:38 am UTC

Derp, that came out badly.
I meant to say that having a script source at the very start of the document (or a link[rel=subresource] as I've just learned) will make the browser aware earlier that it should load that script. And that together with defer it won't block the html parser any further. (and on top of that still execute the scripts in order once the html (and CSS if I recall correctly) parsing is done.)

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Mon Apr 09, 2018 7:23 am UTC

I've applied some JavaScript trickery that will remove the background image and also make the pop-ups on the links work in Internet Explorer. I've tested it in Internet Explorer 6, and it works there. Not sure whether it works in, for example, Internet Explorer 10, can you check it?
Almost all of the scripts now work in Opera also.
The Android Browser (on my Samsung Galaxy S3 Mini) still renders my website incorrectly, and I can't find out why. I'm quite sure there is something wrong with the CSS, but I can't find what.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3502
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: An amateur has tried to build a website in raw HTML

Postby Soupspoon » Mon Apr 09, 2018 8:12 am UTC

I don't know how 'incorrectly' my Android Chrome browser renders things*, but I never did originally notice the background image and had to extract the url from the source to appreciate it at all. A bit of a waste.

I don't think, last time, the left and middle columns independently scrolled (is the fact that it does now something to do with the .js trickery you mention adding? Or did I just not notice/finger-scrolled in a different way last time?) but it's off-putting when trying to drag-scroll the full page (you have to drag the sub-pane to its limit before the full thing scrolls).

That bouncy button over the bottom of the right pane still doesn't seem to do anything useful (just an anchoring link to somewhere already within easy reach on the document) but obscures text. The bouncing (that I've yet to read the code behind in detail) reveals most of the words at some point, but making only when you hit the scroll-buffers do you get it moving (temporarily, like at the moment of the screenshot I've included below) out of the way of the "understanding of how"/"computers work. I myself try" lines. Distracting. Makes it looks far more important than it is, functionally. (IMO. If I haven't missed some other reason for it.)

I will give this another (another) look when I get to a desktop PC. None of the above is 'wrong' (except the obscured text, and the background image adds nothing to my experience but a bit of borderside detailing that I might not miss) but I'm wondering if you're trying to put too many visual tricks into the front page. Less is more. Or, if you wish, "the appearance of less, albeit achieved through complex background tricks, is more", though I also happen to like only just enough scripting/etc to achieve an effect.

You're wanting to showcase your abilities and imagination, I know, so I accept a bit of clutter to draw the eye to things, but I actually was so distracted by the colours, the first time round, that when I tried the main2/main3 alternate style links it would flash on reloading the page and I didn't even realise that the middle segment changed colour from black (styles 1 and 2) to light grey (style 3), or the other more subtle things. I only saw what happened after I looked at the source, and realised what I should have noticed first time round. That may just be my own inobservency, though. ;)


*
Spoiler:
Screenshot_2018-04-09-08-35-01.png
Last edited by Soupspoon on Mon Apr 09, 2018 8:16 am UTC, edited 1 time in total.

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Mon Apr 09, 2018 8:16 am UTC

this is my IE11 at work:
1.jpg


and this is in compatibility view :/
2.jpg

User avatar
Flumble
Yes Man
Posts: 2023
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby Flumble » Mon Apr 09, 2018 1:53 pm UTC

speising wrote:this is my IE11 at work:

What the- how- oh dammit! Please remove any and all IE11 in a 500 kilometre radius.

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Mon Apr 09, 2018 5:14 pm UTC

Flumble wrote:
speising wrote:this is my IE11 at work:

What the- how- oh dammit! Please remove any and all IE11 in a 500 kilometre radius.

well it is the latest native browser on win 7, which is still the last useable win version.

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Mon Apr 09, 2018 5:28 pm UTC

Flumble wrote:
speising wrote:this is my IE11 at work:

What the- how- oh dammit! Please remove any and all IE11 in a 500 kilometre radius.

Thanks, the same solution works in the Android Webkit browser.

User avatar
hotaru
Posts: 1040
Joined: Fri Apr 13, 2007 6:54 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby hotaru » Mon Apr 09, 2018 8:26 pm UTC

speising wrote:well it is the latest native browser on win 7, which is still the last useable win version.

it's not usable anymore:
Spoiler:
Image

Code: Select all

factorial product enumFromTo 1
isPrime n 
factorial (1) `mod== 1

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Mon Apr 09, 2018 8:40 pm UTC

works still pretty well, though.

EvanED
Posts: 4330
Joined: Mon Aug 07, 2006 6:28 am UTC
Location: Madison, WI
Contact:

Re: An amateur has tried to build a website in raw HTML

Postby EvanED » Tue Apr 10, 2018 12:59 am UTC

I also don't think it's fair to say that Win 7 isn't usable any more. Even if you took a liberal definition of "unusable" to include that it's not safe to use because it doesn't get security updates, there's still almost two years left of usability.

User avatar
hotaru
Posts: 1040
Joined: Fri Apr 13, 2007 6:54 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby hotaru » Tue Apr 10, 2018 1:46 am UTC

EvanED wrote:I also don't think it's fair to say that Win 7 isn't usable any more. Even if you took a liberal definition of "unusable" to include that it's not safe to use because it doesn't get security updates, there's still almost two years left of usability.

only if you're paying microsoft for a support contract, since microsoft often considers security updates "new features" and uses that as an excuse to exclude versions of windows that are out of mainstream support from fixes.

Code: Select all

factorial product enumFromTo 1
isPrime n 
factorial (1) `mod== 1

EvanED
Posts: 4330
Joined: Mon Aug 07, 2006 6:28 am UTC
Location: Madison, WI
Contact:

Re: An amateur has tried to build a website in raw HTML

Postby EvanED » Tue Apr 10, 2018 4:34 am UTC

Can you link to an example? (Of a "new feature" that's really a security update)

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Tue Apr 10, 2018 7:01 am UTC

I also have some sympathy for older OS-es. I think that modern OS-es (as well as many other programs) are simply over-designed. They have a bunch of features almost nobody uses, but they slow down the computer. How come Windows XP can run on 64MB of RAM, while Windows 7 barely run on 512MB of RAM? They feel almost the same.

Anyway, does my website work in IE11 now?

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Tue Apr 10, 2018 11:02 am UTC

yes it does! :mrgreen:

User avatar
hotaru
Posts: 1040
Joined: Fri Apr 13, 2007 6:54 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby hotaru » Tue Apr 10, 2018 1:14 pm UTC

EvanED wrote:Can you link to an example? (Of a "new feature" that's really a security update)

how about their failure to add support for any reasonably secure (ECDHE and AEAD) TLS cipher suites to windows vista?

FlatAssembler wrote:How come Windows XP can run on 64MB of RAM, while Windows 7 barely run on 512MB of RAM?

XP and 7 both run very poorly (almost unusable) with anything less than 1 GB of RAM. the only difference in RAM requirements in newer versions is that Microsoft decided to stop lying about how much RAM windows actually needs to run properly.

Code: Select all

factorial product enumFromTo 1
isPrime n 
factorial (1) `mod== 1

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3502
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: An amateur has tried to build a website in raw HTML

Postby Soupspoon » Tue Apr 10, 2018 1:31 pm UTC

My 2K machine is running nicely on 64MB (was originally a '98 box, I think, and I didn't do too much to its hardware…)

Not a point, just thought I'd mention it. (My '95 machine has had 40MB installed almost from the day I bought it new (it came with 8, and I added 32 and a second HDD for *oomph*, and boy did it oomph!)

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Tue Apr 10, 2018 6:52 pm UTC

yes it does!

Is the bottom left SVG gradient corner also rendered properly in IE11? How about the SVG PacMan I've made to see how well it will run in various browsers?
XP and 7 both run very poorly (almost unusable) with anything less than 1 GB of RAM. The only difference in RAM requirements in newer versions is that Microsoft decided to stop lying about how much RAM windows actually needs to run properly.

What are you talking about? Windows XP and Windows 7 need 64 MB and 512 MB of RAM only to install. Once they are installed, they require less than half of those amounts to boot.
Have you tried them in a virtual machine?
Windows XP, once it's installed, can boot with 25MB or RAM in normal mode, and with 19MB of RAM in Safe Mode. If you try to run them in a virtual machine with less than that, you get a "Bad System Configuration Info" BSOD. The Windows XP boot loader fails on a virtual machine with less than 9MB of RAM.
Windows 7 require 234 MB of RAM to boot normally, and 194 MB of RAM to boot in Safe Mode (again, before you get a BSOD during booting). The Windows 7 boot loader requires 35 MB of RAM for the booting to even begin.
Windows 7 definitely requires a few times more RAM to do the same thing Windows XP does.

User avatar
hotaru
Posts: 1040
Joined: Fri Apr 13, 2007 6:54 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby hotaru » Tue Apr 10, 2018 7:10 pm UTC

FlatAssembler wrote:What are you talking about? Windows XP and Windows 7 need 64 MB and 512 MB of RAM only to install. Once they are installed, they require less than half of those amounts to boot.
Have you tried them in a virtual machine?
Windows XP, once it's installed, can boot with 25MB or RAM in normal mode, and with 19MB of RAM in Safe Mode. If you try to run them in a virtual machine with less than that, you get a "Bad System Configuration Info" BSOD. The Windows XP boot loader fails on a virtual machine with less than 9MB of RAM.

I'm not talking about simply booting. I'm talking about actually running useful software once it's booted, and yes, I have tried them in a virtual machine. they both run about the same (acceptable performance) with 1 GB, and about the same (barely usable) with 512MB.

Code: Select all

factorial product enumFromTo 1
isPrime n 
factorial (1) `mod== 1

speising
Posts: 2266
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: An amateur has tried to build a website in raw HTML

Postby speising » Wed Apr 11, 2018 11:23 am UTC

FlatAssembler wrote:yes it does!Is the bottom left SVG gradient corner also rendered properly in IE11? How about the SVG PacMan I've made to see how well it will run in various browsers?

look like in chrome, but there are interesting circle segments which i don't see there:
Capture.JPG

Capture2.JPG

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Wed Apr 11, 2018 6:21 pm UTC

I'm talking about actually running useful software once it's booted, and yes, I have tried them in a virtual machine. they both run about the same (acceptable performance) with 1 GB, and about the same (barely usable) with 512MB.

I am not sure what you mean. Under Windows XP, both Firefox and Notepad++ run without problems on a virtual machine with around 150 MB of RAM. What more do you need?
Look like in chrome, but there are interesting circle segments which i don't see there.

OK, I've tried to add some JavaScript trickery to remove those circles under Internet Explorer. Does it work?
Damn, I am not sure I even could be a professional programer, the computers can be so irritating!

User avatar
Flumble
Yes Man
Posts: 2023
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby Flumble » Wed Apr 11, 2018 10:24 pm UTC

What's the point of those circles in the first place? You're not supposed to see them, right?


Also (assuming you want to stay with a free webhost) you may want to move your site over to GitHub Pages. It's a slight upgrade in terms of advertising and analytics (namely github doesn't inject anything into your pages) and uptime, but more importantly it introduces you to git, one of the popular version control systems.
And knowing how to use a VCS is very important for working in a team, for working alone and even for working on things outside of software development.

User avatar
hotaru
Posts: 1040
Joined: Fri Apr 13, 2007 6:54 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby hotaru » Thu Apr 12, 2018 1:00 am UTC

FlatAssembler wrote:Under Windows XP, both Firefox and Notepad++ run without problems on a virtual machine with around 150 MB of RAM. What more do you need?

https://www.mozilla.org/en-US/firefox/5 ... uirements/
512MB of RAM / 2GB of RAM for the 64-bit version

that's quite a bit more than 150 MB just for firefox.

Code: Select all

factorial product enumFromTo 1
isPrime n 
factorial (1) `mod== 1

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Thu Apr 12, 2018 3:26 am UTC

What's the point of those circles in the first place? You're not supposed to see them, right?

CSS doesn't support conical gradients, and neither does SVG. I've tried to make one by using an SVG with a polygon (made of triangles of different colours) and a circle behind it (so that the "tips" of those triangles don't look transparent when there is an image behind them). That seems to work properly in all browsers except Internet Explorer.
that's quite a bit more than 150 MB just for firefox.

I am quite sure Firefox 43 (the latest version which can be easily installed on Windows XP) can be installed (not just run once it's installed) on a virtual machine with around 150MB of RAM running Windows XP.

User avatar
Xanthir
My HERO!!!
Posts: 5311
Joined: Tue Feb 20, 2007 12:49 am UTC
Location: The Googleplex
Contact:

Re: An amateur has tried to build a website in raw HTML

Postby Xanthir » Thu Apr 12, 2018 8:20 am UTC

FlatAssembler wrote:
CSS doesn't support conical gradients, and neither does SVG.

Note that conic-gradient() is defined in Images 4, was cleared for shipping by the CSSWG, and is implemented in Chrome now (in nightlies).
(defun fibs (n &optional (a 1) (b 1)) (take n (unfold '+ a b)))

User avatar
hotaru
Posts: 1040
Joined: Fri Apr 13, 2007 6:54 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby hotaru » Thu Apr 12, 2018 4:15 pm UTC

FlatAssembler wrote:I am quite sure Firefox 43 (the latest version which can be easily installed on Windows XP) can be installed (not just run once it's installed) on a virtual machine with around 150MB of RAM running Windows XP.

nope: https://www.mozilla.org/en-US/firefox/4 ... uirements/

also: https://en.wikipedia.org/wiki/Firefox_v ... through_49
End-of-life 43.0.x product line on January 26, 2016.


and mozilla is ending all support for firefox on windows xp and vista this year: https://support.mozilla.org/en-US/kb/en ... -and-vista

Code: Select all

factorial product enumFromTo 1
isPrime n 
factorial (1) `mod== 1

FlatAssembler
Posts: 43
Joined: Fri Oct 27, 2017 7:42 pm UTC

Re: An amateur has tried to build a website in raw HTML

Postby FlatAssembler » Fri Apr 13, 2018 6:15 am UTC

Note that conic-gradient() is defined in Images 4, was cleared for shipping by the CSSWG, and is implemented in Chrome now (in nightlies).

That certainly won't work in Internet Explorer 11. Besides, I would have to rewrite almost all the scripts on my website to implement that.
nope:

That's a "recommended hardware configuration". It doesn't really mean anything. It can't possibly take 512 MB of RAM to install Firefox when Firefox itself is 200 MB large. It takes 512 MB of RAM to install Windows 7, which are 16 GB large. Try it in a virtual machine, I distinctly remember I did it (installed Firefox 43 on a virtual machine with around 150 MB of RAM running Windows XP) without running into problems.
And Mozilla is ending all support for Firefox on Windows XP and Vista this year.

Yes, if you try to run the newest version of Firefox on Windows XP, you get a BSOD.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3502
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: An amateur has tried to build a website in raw HTML

Postby Soupspoon » Fri Apr 13, 2018 10:25 am UTC

Linescan a set of orthagonal 1px gradient lines, 1px 'apart', from (in whatever orientation) circumference point to the radial line perpendicular to the scanning. You can work out that the gradient of each line is based on the colour-gradient expected at that point on the radius, the end-point needs the shade needed to maintain constant shade down the radii (you could make an exception for centre-point pixel of indeterminate hue, if you draw that, but it only gets drawn once, so might not be obvious).

It can even be done with purely integer maths, which is how you'd typically do it in assembler-level code (which you might have already written, given your expertise, when you might have to push directly to video memory) though you do have trig functions to make things easier. And you have the line-hue gradient so you don't have to reduce it to finely-chopping the line into 1px lengths, as well as width.

Visually: I mean a bit like this:

Code: Select all

Col1 < - - - - - - - - - > Col1?
ColA < - - - - - - - - - > Col2
ColB < - - - - - - - - - > Col2
ColC < - - - - - - - - - > Col2
  ColD < - - - - - - - - > Col2
  ColE < - - - - - - - - > Col2
    ColF < - - - - - - - > Col2
      ColG < - - - - - - > Col2
        ColH < - - - - - > Col2
          ColI < - - - - > Col2
              ColJ < - - > Col2

(Working out where ColA…J lie on the Col1…2 scale is left as an exercise.)

You can even apply (easily) to a full circle of wrap-around hue-gradienting, or exact-quarters, and even with end-radii of diagonal natures, by further line limiting the relevant orthagonals against the angle-line that represents the radial segment.

Slower than API/system/hardware-level handling, but probably quick enough if you still feel the effect to be absolutely necessary.


Return to “Coding”

Who is online

Users browsing this forum: No registered users and 6 guests