Jump to content

BenoitRen

Member
  • Posts

    977
  • Joined

  • Last visited

  • Donations

    0.00 USD 
  • Country

    Belgium

Posts posted by BenoitRen

  1. I know about STRONG tag semantic value, but why they didn't just "assigned" this semantic value to B (and corresponding one to I tag)? To bloat code of web pages?

    You should read up on the browser wars of the second half of the 90s, where non-semantic elements like b and i were introduced by web browsers. It's not the W3C's fault.

    No, the browser doesn't associate data to their place in the table like would do a spreadsheet program (excell) for example.

    Graphical web browsers don't. However, machines that extract semantic meaning from HTML documents, like search engines, do.

    I agre that there has been abuses with the <table> tag. But that's because there are no <column> tags. So the table tags are used to create columns.

    You're still missing the point of HTML, which is a mark-up language. A column element would be purely presentational.

    It's now advertised to use <div> to create content areas, ok but I'll object that <div> were not intended to create columns neither, but page divisions.

    You turn those page divisions into columns through CSS. That's perfectly fine.

    If the goal is to create columns, <div> + css is not easier nor more intuitive than <table>. JMO

    It's not more intuitive, but it sure makes more sense and makes for less mark-up.

  2. And in what the two uses are different? Displaying lists of numbers or names in a table is... still and only, a form of layout.

    The two uses are different because tabular data is a form of data association. Each row and column is related in some way.

    If you use it for lay-out, there is no such association. Each cell contains either some paragraphs, images, or an unrelated spacer GIF image.

    Allas, the days when the only two tags left to be used legaly in html will be <div> and <span> are closer and closer.

    Total nonsense. Especially because they have zero semantic meaning.

  3. When I plug an USB flash drive into my Windows 95 box, it only takes two seconds for it to be recognised and usable. On my Windows XP laptop, it takes at least 10 seconds.

    Heh, Windows 7. I've heard similar stories from early adopters before Vista was released. Just wait until Windows 7's retail version hits their hard drives.

  4. But then again, you are using Windows 98 in 2009, you probably haven't heard of managed code, so what he says may indeed seem logical to you.

    Know your history. Managed code has been around for a long time. I know Visual Basic and Java were at the time.

  5. Why don't you use tables?

    Because the table element's semantic meaning is tabular data. It should not be used for lay-out.

    This is a M.Y.T.H.!

    Because the specificity of the web is that content is tied to the way it's presented and vice-versa, the presentation influences the meaning of the content, especialy for text.

    This is wrong. HTML is a semantic language.

    flexible and where you want with absolute positioning

    Now this is a myth.

    You can do smart things with tables and old html tags.

    Which inevitably leads to bloated mark-up.

    For example I'v read that the new scholars don't like the <b> tag and prefer the <strong> tag.

    This is INSANE!

    The b element has no semantic meaning.

    And why dropping the <br> tag? I don't understand.

    The most fanatical drop it. It doesn't have to be dropped, though.

    If you look at the source of most famous websites, there are ridiculous amount of garbage in the shape of javascript but also useless pictures, redudant menus, things absolutely useless to read, ads, etc compared to realy useful content.

    This has nothing to do with separation of content and style.

    In the case of Benoit Ren website: SeaMonkey, I realy don't see the advantage of <div> over <table>

    It's more flexible, and less mark-up. I can decide tomorrow that I'd want to have the menu be a bar at the top, for example, without changing any HTML.

    Finaly, one more point in favor of tables vs. separate style sheets: When you save the webpage as "html only" and all the "tables" are on a separate css file, good luck to read it later!

    It will still be readable.

    Chosing <strong> instead of <b> as a standrad is insane for the very reason that it takes 5 characters to type more, 10 with the closing tag.

    There is more to the strong element than just being longer to type.

    And again I reiterate that "I'm positive that a website will look fine on IE6 even if never tested on this browser." And I'm right as demonstrated by BenoitRen's example above.

    Except my design is very simple, and while IE6 doesn't outright break the site, it's not a pretty sight.

    I can't stress enough how small sites with home pages and simple navigation, without database functionality or any fancy stuff, are so very outdated.

    Now, now, this isn't necessarily true. Less is more, and content is king.

    However I don't think it's so important to check the layout and eventualy, themes, transparency, shadow and other cool stuffs in IE6 because that won't break the readability or the original functionality of the website.

    I agree, but people are being paid to make a website that looks the same everywhere.

  6. I'm also wondering why developers keep testing websites on IE6

    Because many users are still browsing with it. We don't want to alienate ~15% of our users.

    I'm positive that a website will look fine on IE6 even if never tested on this browser.

    In practice, this isn't always true.

    Yes and problems arise when you try to do bells and wisthles in css.

    Not just bells and whistles, as I pointed out.

    On many websites the only interractive stuffs seem to be adds.

    I don't think you're looking hard enough. There's a lot of interactivity, even on this very board.

    And... what "min-height" was used for?

    To make sure that my second column stretches to the bottom of the viewport, so the site appears as two columns instead of one column and a box with content (mostly only a problem with the main page. It just looks wrong otherwise.

    The mouseover menu is pure CSS using the :hover pseudo-class intelligently.

  7. I was talking about IE6 vs IE7/8.

    Codes uncompatible with IE7 or above have little use and and are largely unused, sometimes unkown.

    Have you ever made a website using CSS? Pseudo-selectors like first-child and last-child are very useful, as is the > selector. None of those are supported by IE6. They were also often unused, because IE6 still had a big market share, and you also had to support IE6.

    The JavaScript DOM is very useful for little animations, interactive features, and web applications. Something as basic as getComputedStyle is not supported by IE6, and has to be worked around.

    The only reason they want want more "interactive" stuffs is advertising. Save for advertisements, the creation of a website is very simple.

    Utter nonsense. I've made interactive stuff, and none of it has been advertising. The same goes for many things. Websites aren't that simple anymore. Just google for complaints about CSS by people who don't understand it.

    It doesn't take 50% more coding time to support IE6. Just simplify the way you desing webpages.

    Bogging down the design is not a good proposition, and proof that IE6 isn't that good. Nevertheless, Phantasy Star Cave's main page is pretty simple 2 column design, yet I had to use a CSS hack to get it to render properly in IE6.

    Another site of mine, SeaMonkey.be, doesn't render properly on IE6 because it doesn't support the CSS min-height property. And that's a very basic design.

    Posting a comment on a blog is an activity that exists for more than 10 years. Why does it has to be all of a sudden not possible on IE6?

    This is about much more than weblog posting, which is handled server-side.

    I urge you to learn about web design instead of making wildly inaccurate statements.

    Maybe it's better not to program a website with so many good-looking features. Keep it simple. Adding fancy features to a website means a lot of self-inflicted pain.

    See above.

    I prefer sites which display fine in Internet Explorer 5.5.

    Good luck, because that had a wrong box model that could lead to serious rendering quirks.

    But somehow state-of-the-art bells and whistles are the pride of an up-to-date website programmer.

    Look, most CSS is not about bells and whistles. It's basic style information.

  8. If I'm allowed to make one last post (because it's been in my head all day while working), I don't believe in forced upgrading. One should be able to try the new version, and if it has things that are useful to you, you should upgrade. If it doesn't bring anything to the table that you need, you should be able to not upgrade.

    The problem with Windows is it always comes with a lot of new features you don't need, with maybe one or two you might need. The best example is the jump from XP to Vista.

    Finally, we wouldn't still be using DOS. People have chosen to run Windows 3.11 of their own volition. They wanted a GUI. People will upgrade if they think it's worthwhile.

  9. Most of what you can configure in Firefox is gratuitous.

    You must not know about about:config.

    Sorry i wasnt too clear what i meant there. when i say "configurable with GPo" that has nothing to do with home users in my mind (yes i know about local GPo but that hardly counts) I use FF almost exclusively My home PCs & work PCs but on the PCs I manage, IE is MUCH easier to keep up to date and manage default settings with Group Policy Object received from the domain.

    Ah, yes. That's one area that Firefox can improve in. There were some tools available for Mozilla (the suite), though.

  10. The NT lovers aren't listening, like usual, just regurgitating their points. I invite them to check older threads, in which the Win9x users have successfully rebutted their points many times.

    And Windows 9x systems have been made infamous for the fact that they crash constantly, randomly need rebooted due to resource leakage or becoming unstable for no apparent reason.

    Which was mostly due to the forced integration of IE.

    It's easy to say "says you", but it's easy to get past local security in NT. There are backdoors, and Microsoft knows about them. But they won't patch them until they're publicly exposed, as usual. A good friend of mine used to be in a hacker group, and can get admin status in ANY NT computer in a matter of seconds. No joke.

  11. The thing that scares me the most about Firefox is - what will happen when the browser does have the vast majority market share? It's not the most secure codebase, and like anything else it'll become a target.

    You seem to be assuming that a program with the most market share is insecure by default and should be avoided. This is not true. Look at Apache.

    Besides, Firefox is already a target.

    I still dont think its possible for FF* to overtake IE* any measurable amount of time. IE will always control a majority of the market since it it the most configurable with GPo and the like which will always make it the bigger target.

    Firefox is so much more configurable than IE. And it's open source. I also doubt most home users are into customising every aspect of their gateway to the web.

  12. Not at all. Like I said, anything that uses the network is affected by buffer overflows and such exploits. That was merely one example.

    Once the malware has entered your system, you have already lost. It's the same on NT.

    They certainly are, be they software, or software running on a specialized box (*all* firewalls are basically software).

    Ooh, so you're trying to argue semantics now. A firewall running on your computer is not the same as running it on your router, which is a specialised computer that is much more capable.

    According to you, perhaps (BTW, Ranum's ideas are quite funny). The rest of us live in the real world, where bad things happen now and then, and have to be fixed. Just like we have hospitals in case you break your leg.

    Except computer security is not like the real world.

    Your own extremely limited view of security seems to rest solely on having network services or not (being featureless), disregarding everything else, especially when the rest is so full of holes that there's practically only air left.

    Again, you're confusing local security with remote security. We know Win9x has no security model, and that's not necessarily a problem.

    Security is a process, not a product. Not having exploitable services is one thing. Practicing security in everyday use is another.

×
×
  • Create New...