Forum Moderators: open
For example, Google could declare:
If you do X, then we will reward you with PR + Y, when Y is a page rank bonus. The bonus does not have to be high - even a tiny (but tangible) increase in rank can make a big difference to a web site, but has negligble detrimental effect on overal search satisfaction for users. For example, if a site with PR = 6.2 all of a sudden beats a site with PR = 6.3 because it did X, then is the search user more than neglibly worse off? I don't think so, as both pages have good and almost equivalent PR. But it is still enough of a difference to a webmaster for them to pay attention.
So what could X be, that would benefit both Google and the Internet as a whole?
How about:
1) Valid XHTML. XHTML is much easier to parse than tag soup, which would allow Google to more reliably analyse page structure. The W3C has be struggling for years to find a way to move people from tag soup to XML. If Google provided even a small bias towards valid XHTML, I bet all of a sudden there would be a huge increase in deployment, and tool makers would feel enourmous pressure to update their wares to generate correct code.
2) Particular META tags. Now I know that tags such as keywords are open to abuse. But not all tags are so likely to be spammy. Useful tags could include:
parent page
categorisation/classification information
language
primary country
Now there may be many more possibilities. But what I am seeking is a discussion of how Google could offer a carrot to bring about widespread adoption of a technology/standard/conformance that would benefit not only Google, but also the Internet itself.
I suppose if you could say to people: if you do xyz it will help to push you up the SERPS on Google, they might understand that rather than saying things like:
"But my site looks fine on IE5.0 - why on earth should I configure it just so that it looks good on Mozilla, Opera, mac, Linux etc. Nobody uses those browsers / platforms yada yada yada..."
Incidentally can anyone who knows of a good paperback which goes through all the necessary steps required to turn html 4.0 transitional into xhtml send me a sticky please?
I know things like <br> becomes <br /> but I want to have a reference to make sure I don't miss anything when I convert my pages.
Thanks.
/Laughs so much out loud/
I still don't see the point. I design and test in/for mozilla, and test in IE. If it works in those, I know I cover 97% of my visitors (excluding spiders). I hope that the other browsers are not so stupid that they STOP their user from seeing sites that other browsers let them see.
Why validate? It seems just like making me jump through hoops for nothing. I'd get really pissed off if Google gave out SERPS for validated sites. [Really really]. I am from the school of thought which says the internet is for communication, and if I communicate my info with my users, then my site is a success.
Where X= no hidden links/hidden text/cloaking etc.
Standards compliance is great, but to move from one markup language to another, there has to be an excellent reason, not just a bias, IMO.
> But my site looks fine on IE5.0 - why on earth should I configure it just so that it looks good on ...
Yes, if it works in IE and Netscape, and not in other browsers, then I think popular opinion says that the other browsers are broken.
> Where X= no hidden links/hidden text/cloaking etc.
Okay, but isn't that like rewarding someone for not being a criminal?
I still don't see the point. I design and test in/for mozilla, and test in IE.
Precisely. Most webmasters don't see the point of standards compliance. But they do see the point of earning extra PR.
HTML can't be that hard for an SE to parse
I doubt that you have tried to write an HTML parser. It is non-trivial. The hardest part is error handling. Just take a look at the parser in Mozilla - it's mind boggling. And even the parser in Lynx is mighty knarly.
Yes, if it works in IE and Netscape, and not in other browsers, then I think popular opinion says that the other browsers are broken.
By that definition, every version of IE apart from the one you are using is broken. Same goes for other platforms. They all render differently, particularly their handling of broken HTML.
Now I fully understand that most webmasters see no point in producing valid HTML. That may be because they perceive no incentive to do so. If they did, then google would not need to provide an incentive. So I am not arguing that it is good for webmasters - instead I am saying that it is good for the Internet itself.
So does anybody else have suggestions for what would be:
- good for the Internet
- easy for Google to test
- likely to be responsive to a PR incentive
It's like a dictatorship. If I'm a dictator, I see my immeadiate wealth and status, and don't really care for the long-term wealth of my country.
Similarly with narrowminded views on browsers.
How many websites got upset recently wen MS fixed one bug in IE but not another, where the one bug was used to deliver specific CSS rules to avoid the other one. So now there is no way to give IE a non-buggy rule set to make it behave.
Just because other car manufacturers don'T put in seatrbealts and their drivers will die first in an accident is NOT a good argument not to put seatbelts in your own.
I appoligise for my far fetchedcomparisons, but it is just so hard to explain the benefits of standards compliance and other "good behavior" matters to pepole who are not prepared to look beyond their own nose...
Otherwise there would be no crime or bad behaviour ;)
90% use IE, 10% use others because they have a particular skill, programming etc. I think that the percentage of home users who use IE is alot bigger, and they are the people I want to target.
My site tries to bring a different culture to the web, and is starting to be a database of drum rhythms. These drum rhythms are played dynamicaly over the web. To show all the info relevent to the rhythm near the player I needed dynamic text. I also needed a set of good sound functions, for which IE's are superb. My site downloads in seconds, but is one of the most dynamic on the net.
Trying to get that cross-browser at the moment is alomost impossible (I am trying), and for people that aren't even interested in what I have to offer.
For interesting, dynamic sites that arn't just full of text and sales, IE has a much more diverse set of functions, that far exceed the other browsers. When they catch up, and the 'general' public starts using them, I'll put more effort into incorporating them.
But for now, noJS noIE gets a basic site and IE gets the dynamic. Little by little it gets crossbrowser, but like I said earlier, it's only for the proffesionals who won't even be interested in my site.
So your saying I've got to make the site so you can see it, even though you don't want to..........
It does concern me that everybody looks at it the wrong way, they day `How can we make standards, blah blah, rewrite billions of pages of perfectly readable HTML`, when they should be saying `How can we make even more forgiving browsers, so webmasters can become even more sloppy even, and still be fine?`.
I don't get why us webmasters should want to make extra work for ourselves to become standards complient because web browser developers are too lazy to make really robust rendering engines.
(Analogy: Microsoft Word - it KNOWS you are making spelling / grammar errors, it may point them out, or even change them for you if they are really obvious, but it doesn't moan about printing them out if you leave them wrong, and your printer is quite happy with it too! I suspect if Word refused to print badly spelt words for `English Standards Compliance` the whole world'd get pissed off!)
I could make it cross-browser again, using gif animations and long sound files, but this makes each rhythm 100Kb, where as the method I use now I can have a thousand rhythms in the same memory, and have lots more control over tempo, soundrate, balance, volume. Plus the viewer doesn't have to wait 5 mins for each rhythm, once the page has loaded each new rhythms is about 100-500 bytes.
So I could make it cross browser compatible, but then nobody would wait for it to load, and I could only fit 50 or so rhythms on my 10Mb of space.
>>cheaper and easier to only see the short term benefits.
I've spent a year full-time on my site, making a LARGE effort, and that's just getting it to work in IE first...
An IE only site may be IE only, but it certainly isn't easy, especially going from <img src="businesscard.jpg"> as my first page to what I have now, in about a year. (from knowing nothing)
I get it to work in IE only its because safe, and easy to apply, and I could get on with the engine of the site. I do provide alternatives for noIE noJS, and swapping functions for more cross browser ones as I go.
I do this to see the short term benefits (if there would be one), and hope that the short term benefits will give me enthusiasm to go on. (getting dropped from the index doesn't help).
I know that in the long run its better for me to go to XHTML, as my site has the potential of being quite a good mobile browser feature.
Not all programmers have the expertise of XHTML, and if Google made a preference to it, isn't that the dictatorite system. That would give to much emphasis on the programmers capabilities, rather than the content of the site.
I do think valid markup is good and people should use it, but I don't think it has much to do with the real-world quality or usefulness of the content of a particular page. In particular, I think it would be unfair to give a boost to, say, valid XHTML 1.1 over valid HTML 3.2: they're both valid, and many people (especially members of that vast majority of people who have other jobs to do and can't spend all their time keeping up with whatever the W3C dreams up every two years) can't afford to go back and re-code everything.
I thought the whole point of the Internet revolution was that you didn't have to be a member of the technology priesthood in order to participate.
Google rarely uses quotes around attributes yet they will reward us for compliance?
Google is all about content. The structure is just used to define the importance of the content. The vast majority of HTML 3.2 files, even those with errors, render properly in all my browsers. I certainly hope that google never messes with something like this.
Vincevincevince> Your point about 'forgiving browsers' is well made but even if all browsers were very forgiving, that doesn't take away from the importance of standards compliance.
U doent haf 2 spelll Inglisch in enny patklier way for evry won too understand itt. But if you stick to standard spelling conventions which everyone follows, all sorts of things become possible... like indexing in alphabetical order for example. Also it makes it a lot easier for people whose first language is not English to understand it.
Does this mean more work for webmasters? No, writing standards-compliant code isn't any more difficult than writing non-standards compliant code.
But it is:
a) future proofed
b) works on all platforms automatically (like WebTV, imode phones etc.) without you ever having to take account of them
Should Microsoft (or whoever makes the leading browser) or W3C influence what becomes a standard and what doesn't? W3C should. Why? Because their primary concern is people who write the code and people who read the results of the code... not decreasing the market share of other browsers.
The other points that were made about Google serving the searching public first and foremost are absolutely right. It won't be Google that introduces this carrot to webmasters to write standards-compliant code, if anyone at all, it will be somebody else.
