1. Advertising
    y u no do it?

    Advertising (learn more)

    Advertise virtually anything here, with CPM banner ads, CPM email ads and CPC contextual links. You can target relevant areas of the site and show ads based on geographical location of the user if you wish.

    Starts at just $1 per CPM or $0.10 per CPC.

<hX> Tags in Tables

Discussion in 'Search Engine Optimization' started by Owlcroft, Apr 18, 2004.

0
  1. #1
    I'm almost sure I have seen this already discussed here, but I can't find that discussion, so, with apologies, I ask anew.

    Are headings--<h1>, <h2>, and so on--at a disadvantage for being read by search engines if they occur in a table? I realize that deep or complex table structures can effectively bury text, making it seem to be much farther down a page than it is, but would simply placing much of a page inside one table, to get a visual border frame, hurt the keyword force of the <hX> elements on that page?

    The page structure would be:

    usual top structure
    <body &c>
    <table &c.>
    <tr>
    <td>
    normal page content, headers, text, and all
    </td>
    </tr>
    </table>
    some tail-end closing stuff
    </body>
    &c.

    Inquiring minds want to know . . . .
     
    Owlcroft, Apr 18, 2004 IP
  2. digitalpoint

    digitalpoint Overlord of no one Staff

    Messages:
    38,333
    Likes Received:
    2,613
    Best Answers:
    462
    Trophy Points:
    710
    Digital Goods:
    29
    #2
    I would guess they can read the headline tags within tables, although I haven't tested it specifically.

    - Shawn
     
    digitalpoint, Apr 18, 2004 IP
  3. Foxy

    Foxy Chief Natural Foodie

    Messages:
    1,614
    Likes Received:
    48
    Best Answers:
    0
    Trophy Points:
    0
    #3
    They do

    It is not a problem, but do not go too deep with the tables eg more than about 3 [just because I've never tested further than that!] :)
     
    Foxy, Apr 19, 2004 IP
  4. Owlcroft

    Owlcroft Peon

    Messages:
    645
    Likes Received:
    34
    Best Answers:
    0
    Trophy Points:
    0
    #4
    Thank you both. I am working on my other chief web site, and re-building it from the ground up.
     
    Owlcroft, Apr 19, 2004 IP
  5. hans

    hans Well-Known Member

    Messages:
    2,923
    Likes Received:
    126
    Best Answers:
    1
    Trophy Points:
    173
    #5
    just very few years back - there have been a few important ( at that time ) SE preferring only the top few lines of a page, i.e. indexing only the first xyz number of characters ..
    usually around 200 or so.

    now that has changed because other SE have become more important.

    i still belief that some of the OTHER SEs might work according to the old rule. for the simple reason of efficiency with their limited HW resources.
    Google caches more and more pages asking the actual search on cached files - full page. Others have no power to do so.

    i have removed most of my tables from content-rich pages for 3 simple reasons:

    1. less code, faster loading pages - still important for the majority of the world being FAR away from DSL but making MORE than 50% of my site visitors
    2. easier correct display in all mobiles and of course faster loading on mobiles as well as most of them use GPRS ( approx 43KB/s ) or even 9600 B/S.
    3. to avoid any POSSIBLE hiding in deeper page areas and expose all texts as clearly as possible to SE

    at present i have only 3 pages using tables - the 3 door pages with plenty of links.
    ( 1 for the domain-root one for each additional language-root )
    But here i have the titles and some keyword rich description free above/before any tables.

    since the SE results are excellent for me - at least that system without tables appears to be excellent.

    whatever we do as web-designers - we may consider the OTHER parts situation

    1. while speed in www is increasing in most countries making crawling easier
    2. the number of existing pages increases as well and may eventually compensate the added value speed
    3. hence the need to cooperate with SE by providing efficient ( slim ) content whenever possible
    because

    there obviously IS a time-out for SE attempting to crawl / index a site.

    the time-out for Yahoo appears extremely short as when submitting a URL i receive timeout-errors even within less than a second.

    it is obvious that NO SE has the resources ( NOR money ) to spend 20 minutes for a single site being on a slow server -
    hence either a site OFFERS
    content fast and efficient or may either be skipped, partially indexed or indexed less frequently or not at all.

    never talked or seen about ( except Y ) - but it is obvious there must be a timeout for all visits a MAX time by bots allowed for a single page/link to be loaded else page/link dumped/skipped.
    WITHOUT such time-out there would be disaster and hanging bots @ G, Y & Co !

    during some of my previous site review attempts i found some servers SOOO slow ( i addition to WRONG links )
    that after repeated run-starts to crawl for link check i aborted after a FEW HOURS ..
    ( i kill the run usually after 3-6 hrs for a single domain with NO timeout or waiting time before next hit from MY side !!
    that means i run the check on my own site across the web within a few minutes - link check running on a european server / site in California )
    a few hrs 8-12 % CPU of a dedicated server on a back bone .. Google never could afford to waste such resources.

    If you consider such facts - by looking at your site through the "eyes" of bots as seen from the OTHER side - from a SE / crawler side ..
    then you may see the importance of making as slim and as fast loading pages as possible on fastest possible servers
    to assure that the bottle-neck of data flow never is on your side.
    and by presenting content and important data as direct, as clean and as free of code as possible to allow max speed in parsing a page
    and to AVOID any possible error in parsing code ... just see your last title-tag error with line feed. such may be an unofficial bug or rule,
    actually may be even a "bug" in G bot - fact is that complex data structures in the presentation and layout MAY AND DIO ( as in your documented case )
    cause problems.

    Murphy's law applies to ALL in all situations of life
    -->> hence ANY eventually possible probable cause for a misinterpretation of data must be avoided to assure that your pages
    are indexed properly even by faulty bots. simply by never offering a challenge to bots but pure clean data to straight forward FEED bots
    they HAVE to absorb ( indirectly ) all data when pages are parsed.

    G gets faster and faster - Y did so too in very recent days and weeks
    now its up to us to contribute OUR share to speed.

    the recent title tag documentation sent by email to you shows that parsing any tag presents a danger even to Googlebot and related
    processing SW @ G!

    may be you would like to share that one problem related to parsing/processing your recent <titlle> tags to help others avoiding same mistake and to make aware of how small details can screw up a page indexing run by Google and thus the page position and finally the $ flow.
     
    hans, Apr 19, 2004 IP
  6. Owlcroft

    Owlcroft Peon

    Messages:
    645
    Likes Received:
    34
    Best Answers:
    0
    Trophy Points:
    0
    #6
    What Hans is referring to was an idiot mistake I made with my site, but one whose consequences are instructive.

    I have a very large number of pages that are re-made anew every night: there is a template that is filled in using AWS data from Amazon. In the course of making a "minor" change in that template, I added a <meta> tag that, through my carelessness, came out in the middle of the <title> tag--so that the <title> tage was effectively rendered garbage.

    I had been, for some pretty competitive keywords, in the low 20s and gaining when that happened; I went almost oernight to the high 30s, and have stayed there, waiting for Google to reindex those pages.

    The moral is that, even on secondary pages, titles are important!
     
    Owlcroft, Apr 20, 2004 IP
  7. hans

    hans Well-Known Member

    Messages:
    2,923
    Likes Received:
    126
    Best Answers:
    1
    Trophy Points:
    173
    #7
    to be precise:

    the following was EXACTLY on that web page ( may be all 10'000+ ?)

    <title>Great Science Fiction and Fantasy, Edition Information:
    "Peter Pan in Kensington Gardens"
    by J. M. Barrie
    </title>

    hence at first glance a correct title-tag - EXCEPT twice an INVISIBLE but still parsed LF/CR ( new lines in the middle of a metatag )

    and this is what Google saw:
    Great Science Fiction and Fantasy, Edition Information: <meta name ...

    Google parsed the <title> tag UNTIL the first line-feed
    then skipped the rest and went on to next meta tag.

    result
    majority of title tag NOT parsed and NOT used for indexing/positioning

    it shows that Google of course loves ONE metatag on ONE line ! WITHOUT carriage return and line feed within meta tag.

    the essence is a substantial loss in popularity of that page (all such pages ) for a while

    and the key point is

    EVERY byte is parsed - visible AND invisible bytes
    ONLY the cleanest CODE gives accurate results when parsed / indexed !
     
    hans, Apr 20, 2004 IP
  8. catanich

    catanich Peon

    Messages:
    1,921
    Likes Received:
    40
    Best Answers:
    0
    Trophy Points:
    0
    #8
    Nice one Hans
     
    catanich, Jul 23, 2008 IP
  9. Bloomtools

    Bloomtools Peon

    Messages:
    1,361
    Likes Received:
    12
    Best Answers:
    0
    Trophy Points:
    0
    #9
    nice .. hans!
     
    Bloomtools, Jul 23, 2008 IP
  10. hans

    hans Well-Known Member

    Messages:
    2,923
    Likes Received:
    126
    Best Answers:
    1
    Trophy Points:
    173
    #10
    guys
    this thread is 4 years old and apparently solved on the website OP
    and as far as I remember Gogle also has learned to correctly parse a LF/CR in a meta tag some times later after this problem has occurred.
    nevertheless that old problem showed ones more that proper code and code verification beyond visible code IS an absolute MUST for all publishers. INVISIBLE characters even nowadays are creating validation errors and parsing errors by major SE - thus resulting in bad SERPS due to invisible errors.
    web publishing more than ever before is a PROFESSION like medical doctor and racing car mechanic - webmaster is more than a mere email address or self proclaimed title - webmaster is web-MASTERING = knowing all from HW to SW to SEO to dataflow inside SE.
    those who learn steadily succeed - all others fight for survival and finally drawn in the flood of hard working / learning and daily increasing competition.
    unfortunaltely even here in DP forum in many "help me" cases and site review requests, a typical superficial answer often is something to the extent of "looks great" but those giving such replies seldom or never look at source code and how machine friendly + validated a page is.
    SEO is more than meta tags and H1 titles, etc - it includes such most basic topics of correct clean character set encoding, and as in OP problem the clean page FREE of invisible characters. invisible characters are like heavy to digest food for a human stomach.
    a salad or mixed vegetable may be easy to digest
    or
    be salmonella intoxicated just like in recent weeks in USA with tomato/pepper ...
    at first glance edible
    when looked thru a microspcope however potentially fatal.

    data "food" presented to SE needs to be machine friendly - machine means CPU needs to be able to create a clean data flow
    such topics are better understood by those who have a basic understanding of ASSEMBLER and logical hardware components.
    whatever goes thru C or C++ or other softare, later on also needs to flow THRU the actual hardware logical components to create a correct output !!!
     
    hans, Jul 23, 2008 IP