Forum Moderators: Robert Charlton & goodroi
Maybe 50 to a 100 pages created in notepad only to rank, get pagerank etc. Inbound and outbound links. inactive links for navigation.
Any thoughts to do with such a website regarding sever, navigation, raking, pagerank etc?
Thanks for any help and ideas.
[edited by: engine at 1:23 pm (utc) on Oct. 2, 2006]
Well, fair enough, I'll be interested to hear the results of your experiment.
trillianjedi,
It is not an experiment, though you could look at it that way. Many elements of it seem very beneficial to a website.
After all, one can create a vector graphics website with textual content within the vector. Search engines can now read inside them and where their links point to etc.
No, not an experiment but a dedicated txt notepad website. Just like any other website but created in notepad only and displayed as created.
[edited by: engine at 1:17 pm (utc) on Oct. 2, 2006]
Maybe I'm missing something here... is this website going to consist only of .txt files?
Patrick,
Yes, created in notepad and presented exactly the same way but on the internet. A website per say.
[edited by: engine at 1:17 pm (utc) on Oct. 2, 2006]
As a matter of interest I think that pages with no outgoing links might well result in wasting Google PageRank, but this may not matter too much if you have the right incoming links.
As a matter of interest I think that pages with no outgoing links might well result in wasting Google PageRank, but this may not matter too much if you have the right incoming links.
Patrick,
I did an anchor link search in google not long ago and noticed a .txt file listed.
I even came across a html website that had a 4 pagerank but also had a txt file of 6 pagerank. Since google reads and acknowledges sitemaps, robots files etc, one has to consider the benefits of a txt file. I will be very surprised if google follows links in a sitemap txt file but does not in any other txt file.
[edited by: engine at 1:17 pm (utc) on Oct. 2, 2006]
The site would be indexed by SEs that's for sure.
The links can only be copy pasted that's also a fact.
What keeps you from doing it?
[edited by: photopassjapan at 12:45 pm (utc) on Oct. 2, 2006]
those pages will be 'dangling' links as far as the algorithm is concerned.
That's an interesting aspect.
I think the other interesting aspect is whether Google will find and index new pages based on the existence in another .txt file page of pure text non-hyperlinked URLs.
If both of the above turn out positively, then I guess the answer is yes, you can have a pure-text based website.
I still think that users ultimately will not thank you for lack of navigation though.
Having said that, I believe it is a fact that more and more web users are relying on SE's as their navigation system. The web, from an SE perspective, appears more and more like a single big website every day.
TJ
Yes, I don't doubt that .txt files can be strongly listed. I meant that by having no links out of any page, there's a PageRank issue in that links to those pages will be 'dangling' links as far as the algorithm is concerned.
Patrick,
If the site is successful, I could create its html version. By default, present the html version to the human eye and present the txt based to the robots.
I'm looking into ways that does not require IP delivery. That would be cheating. I don't want to cheat. It is my website and I have the right to present what I want to whomever I want.
After all, millions of websites have duplicate websites. Some triple versions. www non www etc. That is cheating, but two versions are not cheating. It all resides in one domain. Not two domains like the www and non www.
Two legitimate versions. Both exactly the same. Possibly a dynamically produced page of the txt file but with some frills and spills of html and css.
There may exist a possibility that that a txt based site may not suffer canonical issues, it may, I don't know yet. If it is safer, I could keep the html version totally out of sight from robots and google.
[edited by: AlgorithmGuy at 1:03 pm (utc) on Oct. 2, 2006]
[edited by: engine at 1:16 pm (utc) on Oct. 2, 2006]
I still think that users ultimately will not thank you for lack of navigation though.
trillianjedi,
I'm thinking on that.
In order for users not to thank me, they will have found the website. Their dissaproval can be interpreted positively.
The user might not thank me but the search engines might.
It is very possible to display a jazzed up version if required and still retain the power of the txt based website.
[edited by: engine at 1:15 pm (utc) on Oct. 2, 2006]
Sure. Go for it and see what happens - that's the only way to find out.
trillianjedi,
But I need to know how a server will dish out the material and what to do to make sure the site works properly.
If a crawler goes to the server and requests http://example.com/ I need to make sure that, because no index.html exists, that the bot is served index.txt
I got to make sure no error pages or restrictions apply. If the host is worth his salt, he will make sure the server responds accordingly by adding index.txt to the configuration defaults.
.
So the issue for me is not whether one thing or another is cheating but whether there's any benefit to it. If you can't reliably introduce any markup into a page, then you can't benefit from page titles, headings, styles, and hyperlinks. Those aspects will have negative ranking implications which may or may not be outweighed by the benefits of the approach.
It seems I was wrong again. It's been pointed out to me (and I subsequently tested it again) that HTML markup in a .txt file is treated differently in IE6 to what it is in Firefox and Opera - all on Windows. FF and Opera treat the markup as plain text. Don't know about the Mac or IE7.
Patrick,
We are not worried about browsers just yet. Humans will see what we want them to see. In all browsers.
Google, googlebots, deepcrawl googlebots and harvesting bots. Maybe throw in mediapartners bots. If the text based website proves successfull, why not, maybe I can cash in on the adsense.
And my text based site should in theory be endorsed by DMOZ for a prized link pointing to index.txt
.
[edited by: Patrick_Taylor at 3:39 pm (utc) on Oct. 2, 2006]
