Beefy Boxes and Bandwidth Generously Provided by pair Networks
Keep It Simple, Stupid

Re: Google "related" search bias?

by themage (Friar)
on Jan 17, 2007 at 13:58 UTC ( #595079=note: print w/replies, xml ) Need Help??

in reply to Google "related" search bias?


I think that this may be a consequence of perl community being so closed and of most of the biggest perl related sites use user unfriendly URLs. The true is that most of us are not web experts, and most don't want to be, so we make funcional web sites, but with very hurtful SEO mistakes.

NOTE: I don't say that we must be Search Engine directed, and that our website users must be put after search engines.

What I'm saying is that won't hurt anyone if some things were taken into account. Some times we are even elitists and/or take uninformed decisions.

Some other great perl resources (like perlmonks.(org|net|com)) don't even allow search engines to crawl and index them.

Add all of that and you have a big perl community that is not indexed by search engines and, therefore, is not taken into account when showing related searchs.

Taking (i like it more), what could be improved (with just a quick look):
  • Clean the robots.txt - or specify only the bad robots
  • use another url form, the node_id=99999 don't help
  • use more css and a lot more HTML to format the site

Mostly, the site structure is pretty nice, it is nevagable, everything we could want is at hand. This is an awful (and wonderfully) big perl site, with everything any webmaster could need.

Some times we look at SEO related topics as a way to improve the money problogger (or so called) earn from their blogs and sites, but it is a way to improve targeted trafic to a site, and that can be "monetized" anyway the site owners like. We can use the extra trafic to take perl to other starting programming, and those who know little perl.

This is a site with pagerank 7 and, as other sites like this, that could help relate perl to a lot of other keywords, including tutorial, sql, web, ...

Time is what we make of it, and even if I have little of that, I can help improve this site (without touching the contents, that are what make this site important and relevant) to make it crawled and more crawlable.

Only making more perl related sites crawlable and more conformant with Search Engines Guidelines we can put more perl in the search results and other relevant keywords in the related searchs.

Replies are listed 'Best First'.
Re^2: Google "related" search bias?
by webfiend (Vicar) on Jan 17, 2007 at 17:40 UTC

    I see that you've thought about this a lot, but I don't get it. Searching for "perl" at yields what I consider the most relevant links on the front page, including a link to Perlmonks. A search for "perl tutorials" puts the Tutorials section of this site right at the top. Google does not seem to have much trouble crawling all over us, despite our URLs.

    How many other keywords would we want to associate Perl sites with? The "related" feature on google is nifty, but I'm not interested in spending any time trying to sort through its inner workings just to get "php" off the list of related searches. It's not my business if people want to look at PHP after looking at Perl. They'll find a good language eventually.

      "perl tutorial" is probably a more common search, and perlmonks is down in the pack for that one.

      A bit of fiddling shows that Google is profoundly ignorant about this site, especially since seems to be defunct.

      A good, fast search engine on this site would be a huge. SuperSearch is a great tool and I use it regularly, but it is slower than Christmas. Last I checked this site is difficult to index properly, so if someone is volunteering to help, I am in favor of it. Not that my opinion matters much!

      Then again, if it was easy to search, anybody could do it. Maybe we are better off being off the map.

      It should work perfectly the first time! - toma

        It's been so long now that I can't remember how I found PerlMonks, but I have a feeling it was through Google. Still, I agree that it would be nice to see more of PM showing up on Google. I don't know if that's our responsibility or theirs. Should we be taking time away from improvement of this site to make it easier for new people to find us? Are people even having that much trouble finding us? How many Monks are there nowadays, anyhow?

        ...SuperSearch is a great tool and I use it regularly...

        You are a better geek than I am :-) I never have managed to get the hang of SuperSearch, so I end up flailing around almost randomly trying to find information, or just digging down through categorized Q & A. I welcome improvements of any kind to SuperSearch! is on the front page of the "perl tutorials" search on Google for me. I don't consider anything on the first page to be terribly low. The results ahead of PM's tutorial also appear to be Perl tutorials, and therefore also relevant.

        Getting a little boost in search engine ranking wouldn't hurt. I would think, though, that a better goal than being seen as more relevant would be to actually be more relevant. As PM's already one of the most relevant Perl sites around, that could take some real effort to pull off.

        I don't see a struggle in site optimization among members of the Perl community as that productive of an exercise in the end, either. It may help us to get awareness out about Perl if the sites were indexed well, but having the best Perl sites rise and fall organically on a search for Perl is probably the best way to handle such targeted searches for the community as a whole.

        Here's some examples of searches that Perl sites might be expected to fall under:
        searchresults pagetype
        programming language 2 direct
        easy programming language 1 as #7
        simple programming language 1 at #8
        bioinformatics 3 O'Reilly conference with "perl" in description
        bioinformatics programming 1 A Perl bioinformatics course is #3.
        artificial intelligence > 10 I only looked through ten, found nothing
        rapid development > 10 same as above
        report generation 9 Consulting firm mentions Perl in site description
        automating system administration 1 IBM has an article on doing this with Perl at #3, while O'Reilly's Perl for System Administration is #9.
        line noise 9 talks about how despite some people find Perl's syntax opaque, it's the world's most flexible programming language
        regular expression 1 Several. Some specifically Perl, some including Perl with other languages.
        CGI 1 MSA is ahead of Lincoln Stein's, but both are on page 1.

        And DBI is the first result overall for 'database interface'.

        Over all, I think Perl is pretty well represented. Python, Java, Ruby, and PHP are often ahead of Perl in those results, but there might very well be more info on the web about those topics in those languages for all I know. I was a little surprised about bioinformatics, report generation, and rapid development being so bereft of Perl-related results.

        Christopher E. Stith

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://595079]
[marioroy]: choroba++, Discipulus++. It depends on the type of module. Data-type "only" modules are likely multi-process safe, re: Hash::Ordered, Tie::IxHash.
[marioroy]: ... when shared via MCE::Share-> share(...)
[marioroy]: Net type modules are likely not multi-process safe unless stated in the documentation.
[marioroy]: The Prima author fixed his module to be both thread and multi-process safe. Thanks Dmitry.
[marioroy]: Of all the GUI-type modules, Prima was the worst regarding thread/multi- process safety. Now, it's the best for safety. ;-)
[marioroy]: Tk, Gtk2, Gtk3 requires extra care.

How do I use this? | Other CB clients
Other Users?
Others contemplating the Monastery: (8)
As of 2017-09-22 10:13 GMT
Find Nodes?
    Voting Booth?
    During the recent solar eclipse, I:

    Results (260 votes). Check out past polls.