|
|
|
|
||||||
| alt.internet.seo Internet search engines and related topics. |
![]() |
|
|
LinkBack | Outils de la discussion |
|
|
#1 |
|
Messages: n/a
Hébergeur: |
Hello,
I've just read Peter Kent's Search Engine Optimization book. It generally seems pretty clear - except for the things most specifically related to my case. (Murphy's Law?) 1. About finding a hosting company, Kent says, "Don't get an account in which you have a subdirectory of the hosting company's domain name." Uncharacteristically, he gives absolutely no supporting reasons. I've had a Geocities account for 10 years. Is there any reason that will interfere my search engine optimization? Do I really need my own domain name? 2. After being www.geocities.com/CapitolHill/Lobby/7049/ for some years, Geocities gave us Yahoo!-based alternate names, mine being www.geocities.com/donaldsauter/ . That seemed much neater and so I started to use it. Now my pages appear both ways in the various search engines. (I don't think I've seen a case of the same web page appearing both ways in one search engine.) Is this working against me? I presume I need to standardize all my urls one way or the other (or get my own domain) before asking others to link to me. If so, is one better than the other? (Is /CapitolHill/Lobby/7049/ the "real" url, and /donaldsauter/ merely some sort of pointer or "logical" name?) 3. Kent emphasizes the importance of the links between your pages on your own site. Until now I have used the short and simple, <A HREF="base8.htm">Base 8 proposal</A>, for example. I see now that, for most of my pages, Google says there are no backlinks. Does that mean it really doesn't count the short form as a real link? For spreading pagerank among my own pages do I need to use the complete url, starting with http:// , on my own site? 4. Kent's book is all about single-purpose web *sites*, I have a wide variety of unrelated web *pages*. It makes no sense to choose one category in the Open Directory Project directory, for example. Is there a strategy for submitting individual pages, or collections of related pages, to appropriate categories within such directories? If I got a bunch of new, free sites, and created an index page in each one that linked to related pages on my main site, could I sneak them all in ODP? While I'm at it, I have a question about a specific page of mine. It was called mykumon.htm and now I'm changing it to kumon.htm (as per Peter Kent's advice). I have seen it as high as no. 18 in Google in a search on the single word "kumon". Then it plummeted down into the 100s. Then it shot back up to no. 19. Then it fell down into the 100s again. Since I now understand the ratings have nothing to do with actual page visits, here is my best guess at this wild behavior. Googlebot comes along; recognizes what a fantastically interesting and useful page it is based on content alone; gives it a good rating; goes on its way to other pages containing the word "kumon" which have little useful content but a good link from www.kumon.com, say, and starts tossing them back above mine. Sounds crazy, I know. Any better explanation? Any chance Google will ever work page traffic and visitor satisfaction into their ratings, before I sink all this effort into diddling with links? It's not clear to me that there would be any more trickery than with this system based on links, or that it would be any more difficult to deal with. Wishful thinking. . . Thanks for your . Donald Sauter http://www.geocities.com/donaldsauter/kumon.htm http://www.geocities.com/CapitolHill...49/mykumon.htm |
|
|
|
#2 |
|
Messages: n/a
Hébergeur: |
"Donald Sauter" <donaldsauter@gmail.com> wrote:
> Hello, > > I've just read Peter Kent's Search Engine Optimization book. > It generally seems pretty clear - except for the things most > specifically related to my case. (Murphy's Law?) > > 1. About finding a hosting company, Kent says, "Don't get an > account in which you have a subdirectory of the hosting > company's domain name." Uncharacteristically, he gives > absolutely no supporting reasons. I've had a Geocities > account for 10 years. Is there any reason that will interfere > my search engine optimization? Do I really need my own > domain name? At about 7 USD/year do you really care? There are several SEO advantages in having your own domain name. > 2. After being www.geocities.com/CapitolHill/Lobby/7049/ for > some years, Geocities gave us Yahoo!-based alternate names, > mine being www.geocities.com/donaldsauter/ . That seemed > much neater and so I started to use it. Now my pages appear > both ways in the various search engines. (I don't think I've > seen a case of the same web page appearing both ways in one > search engine.) Is this working against me? Yes, it's called duplicated content. > I presume I need > to standardize all my urls one way or the other (or get my own > domain) before asking others to link to me. If so, is one > better than the other? which one has keywords in the URL that are related with content? > (Is /CapitolHill/Lobby/7049/ the "real" > url, and /donaldsauter/ merely some sort of pointer or "logical" > name?) If both give exactly the same content, ie, if you change a page, they are both updated, then one is internally redirected to the other. > 3. Kent emphasizes the importance of the links between your pages > on your own site. Until now I have used the short and simple, > <A HREF="base8.htm">Base 8 proposal</A>, for example. I see > now that, for most of my pages, Google says there are no > backlinks. Does that mean it really doesn't count the short > form as a real link? For spreading pagerank among my own > pages do I need to use the complete url, starting with http:// , > on my own site? Every user agent (browser, bot) turns the short form into the long form. This is needed for the protocol that communicates with the web server (it's a bit more complicated, the host name is ripped off, and added in a separate header unless you use a really old user agent, or one that doesn't support the Host header). > 4. Kent's book is all about single-purpose web *sites*, I have > a wide variety of unrelated web *pages*. So do I: http://johnbokma.com/ Between 12,000 and 14,000+ visitors. A day. Maybe Kent should have put that book where the sun doesn't shine, or ask me for a technical review or two. > It makes no sense to > choose one category in the Open Directory Project directory, > for example. Is there a strategy for submitting individual > pages, or collections of related pages, to appropriate categories > within such directories? Yes, just do it. http://johnbokma.com/perl/ is in DMOZ, under Perl of course (or Perl scripts, can't recall). > If I got a bunch of new, free sites, > and created an index page in each one that linked to related > pages on my main site, could I sneak them all in ODP? If it's the "Look here what I found" type of page, probably not. > Any chance Google will ever work page traffic and visitor > satisfaction into their ratings, Doubtful. Both can be bought too easily. > before I sink all this > effort into diddling with links? It's not clear to me that > there would be any more trickery than with this system based > on links, or that it would be any more difficult to deal with. > Wishful thinking. . . Good content and semantic mark up are the keywords. If you didn't find those mentioned at least 40 times in that book, use it to keep you warm or to light a BBQ. -- John Need with SEO? Get started with a SEO report of your site: --> http://johnbokma.com/websitedesign/seo-expert-.html |
|
|
|
#3 |
|
Messages: n/a
Hébergeur: |
On 13 Nov 2006 19:30:18 -0800, "Donald Sauter"
<donaldsauter@gmail.com> wrote: >Hello, > >I've just read Peter Kent's Search Engine Optimization book. >It generally seems pretty clear - except for the things most >specifically related to my case. (Murphy's Law?) > >1. About finding a hosting company, Kent says, "Don't get an >account in which you have a subdirectory of the hosting >company's domain name." Uncharacteristically, he gives >absolutely no supporting reasons. I've had a Geocities >account for 10 years. Is there any reason that will interfere >my search engine optimization? Do I really need my own >domain name? Yes. Geocities is amateur night from an SEO POV. Can you do a 301 from your Geocities domain to your hypothetical new one? I doubt it. There goes all your provenance. You have to begin all over again. >2. After being www.geocities.com/CapitolHill/Lobby/7049/ for >some years, Geocities gave us Yahoo!-based alternate names, >mine being www.geocities.com/donaldsauter/ . That seemed >much neater and so I started to use it. Now my pages appear >both ways in the various search engines. (I don't think I've >seen a case of the same web page appearing both ways in one >search engine.) Is this working against me? You should do one or the other, not both. > I presume I need >to standardize all my urls one way or the other (or get my own >domain) before asking others to link to me. If so, is one >better than the other? (Is /CapitolHill/Lobby/7049/ the "real" >url, and /donaldsauter/ merely some sort of pointer or "logical" >name?) Getting the hell away from Geocities is the best idea. Never going there in the first place, now you're talking. >3. Kent emphasizes the importance of the links between your pages >on your own site. Until now I have used the short and simple, ><A HREF="base8.htm">Base 8 proposal</A>, for example. I see >now that, for most of my pages, Google says there are no >backlinks. Does that mean it really doesn't count the short >form as a real link? For spreading pagerank among my own >pages do I need to use the complete url, starting with http:// , >on my own site? It would probably to use the full url. I don't think you've quite got your head round this but it would probably . >4. Kent's book is all about single-purpose web *sites*, I have >a wide variety of unrelated web *pages*. It makes no sense to >choose one category in the Open Directory Project directory, >for example. Is there a strategy for submitting individual >pages, or collections of related pages, to appropriate categories >within such directories? If I got a bunch of new, free sites, >and created an index page in each one that linked to related >pages on my main site, could I sneak them all in ODP? Themed sites are easier for the engines to deal with, it's clearer to them what they're about. Also, scratching your head over what will or what won't work with the ODP is largely a waste of time as nothing works, it's an out of date idea that is on its last legs. Worry about getting into Digg or um, sites like that (!) >While I'm at it, I have a question about a specific page of mine. >It was called mykumon.htm and now I'm changing it to kumon.htm >(as per Peter Kent's advice). I have seen it as high as no. 18 in >Google in a search on the single word "kumon". Then it plummeted >down into the 100s. Then it shot back up to no. 19. Then it >fell down into the 100s again. Since I now understand the ratings >have nothing to do with actual page visits, here is my best guess >at this wild behavior. Googlebot comes along; recognizes what a >fantastically interesting and useful page it is based on content >alone; gives it a good rating; goes on its way to other pages >containing the word "kumon" which have little useful content but >a good link from www.kumon.com, say, and starts tossing them back >above mine. Sounds crazy, I know. Any better explanation? Don't change the title, it'll be treated as a brand new page with no provenance. >Any chance Google will ever work page traffic and visitor >satisfaction into their ratings, before I sink all this >effort into diddling with links? How would they assess it? You'd have robots all over the web, clicking through to pages and staying there for ages to pretend the content is so absorbing. > It's not clear to me that >there would be any more trickery than with this system based >on links, or that it would be any more difficult to deal with. >Wishful thinking. . . It's clear to Google :-) BB |
|
|
|
#4 |
|
Messages: n/a
Hébergeur: |
Donald Sauter wrote:
> Hello, > > I've just read Peter Kent's Search Engine Optimization book. > It generally seems pretty clear - except for the things most > specifically related to my case. (Murphy's Law?) > > 1. About finding a hosting company, Kent says, "Don't get an > account in which you have a subdirectory of the hosting > company's domain name." Uncharacteristically, he gives > absolutely no supporting reasons. I've had a Geocities > account for 10 years. Is there any reason that will interfere > my search engine optimization? Do I really need my own > domain name? Those types of names characterizes your site as a personal or amateur web site. Is that how you want to be know as? A Mr. No-Name web site? Why are you even asking such silly question? Try reading Mr. Kent's book over again, please! |
|
![]() |
| Outils de la discussion | |
|
|