Re: SQL2005: shred the feeds or store as XML?

Tech Tip: Click here to run a free scan for Windows Errors and optimize PC performance

My blog itself does not give you the answer :-). But you may want to follow
the whitepapers.

Also note that much of the RSS/Webfeeds is a more recent development. I
actually gave a nice demo during my SQL Server launch presentations in
Switzerland where I build an RSS feed over some bug tracking data. I will
hopefully get to post it soon after the holidays...

Basically, if the information is mainly "relational" in nature, gets changed
often and repurposed for other means, you may be better off with having it
shredded.

If you want to keep the original message around for faster processing or for
keeping it around for "auditing", you may want to keep it together. Or, you
may want to do both under certain circumstances....

Best regards
Michael

"clintonG" <csgallagher@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:%23WjuInmBGHA.2664@xxxxxxxxxxxxxxxxxxxxxxx
> Thanks Michael and Hilary.
>
> The first thing I did was review Michael's blog [1] and observe much in
> general that is going to prove insightful but nothing, nada, zip, zero, as
> in where's da' beef with discussions I'm sure you'll both agree is rapidly
> becoming fundamental to the way content is created and transported across
> the web; syndicated web feeds? Lots of big picture stuff going on for
> sure.
>
> I'll thank Microsoft for finally coining a useful and understandable term
> "WebFeed" and note the WebFeed [rss|xml] files are being kept in the file
> system where they can be accessible to aggregators and readers using HTTP
> as well as developers using APIs but to provide a searchable interface,
> support analytics and other benefits to those whose WebFeeds are being
> hosted shredding elements into relational tables is I have to conclude a
> "must do" process. Along comes SQL 2005 and the whole WebFeed can be
> stored in a table and I'm frozen with indecision.
>
> Since WebFeeds have been around since the time when SQL 2005 was first
> being planned and are about to become commodities I am still astounded
> that there is nothing specific that Microsoft has built into the box or
> has been publishing and it again, specifically, it sure seems .NET
> bloggers are mum or have not coalesced at this time.
>
> To respond to Michael's and Hilary's replies, I would say as a matter of
> context, feed items are either being modified frequently as certain
> vertical markets may change price lists, special offers and so on which is
> content if a feed item itself. A given channel will be modified rarely.
>
> Others such as bloggers are adding gruesome chunks of content on a
> whimsical basis. The chunks themselves are rarely if ever modified but one
> or more comments may be appended to a chunk at any given arbitrary point
> in time. Don't ya' just like a sure thing :-)
>
> These are my observations and others are needed. Refactor my insights
> please :-)
>
> <%= Clinton Gallagher
>
> [1] http://sqljunkies.com/weblog/mrys/
>
>
>
> "Michael Rys [MSFT]" <mrys@xxxxxxxxxxxxxxxxxxxx> wrote in message
> news:eK1x5teBGHA.3984@xxxxxxxxxxxxxxxxxxxxxxx
>> Another "rule" is whether you often repurpose part of the data in
>> different ways (shred it) or you mainly retrieve the full XML document
>> (keep it whole). And then you may as well promote properties or define
>> indices for faster access...
>>
>> Best regards
>> Michael
>>
>> "Hilary Cotter" <hilary.cotter@xxxxxxxxx> wrote in message
>> news:%23Mpoy7dBGHA.2568@xxxxxxxxxxxxxxxxxxxxxxx
>>>I think the "rule" is that if you need to update the data frequently you
>>>get better performance if you shred it. Otherwise store it as XML in the
>>>XML data type.
>>>
>>> --
>>> Hilary Cotter
>>> Looking for a SQL Server replication book?
>>> http://www.nwsu.com/0974973602.html
>>>
>>> Looking for a FAQ on Indexing Services/SQL FTS
>>> http://www.indexserverfaq.com
>>>
>>> "clintonG" <csgallagher@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote in
>>> message news:ORLQ%23BcBGHA.3840@xxxxxxxxxxxxxxxxxxxxxxx
>>>>I hope I can get some insight into this issue. I'm developing a feed
>>>>builder and intend to host feed data to make it available for search and
>>>>analytics and so on. What's my best strategy? Shred the feeds or use
>>>>SQL2005 to store the whole file as XML? Any white papers or other
>>>>documents about this specific concern?
>>>>
>>>> <%= Clinton Gallagher
>>>>
>>>
>>>
>>
>>
>
>


.

Relevant Pages

  • Re: ERROR - SQL Server 7, Reporting Services, and SQL 2000 Functions
    ... My .Net blog, ... > SQL7 did not support collations like SQL2000 does and the DATABASEPROPERTY ... > Kanatype sensitivity = false ... > Jasper Smith (SQL Server MVP) ...
    (microsoft.public.sqlserver.server)
  • Re: Has the RSS/Blog vine become overloaded with redundant references?
    ... Feeds get indexed, aggregated, propagated, re-used, manipulated and so forth. ... Some blog posts in themselves are RSS feeds as to allow tracking by the ... > like Digg rate stories the Feed Filter can ... > are my smart filtering options now and what do you see happening in the ...
    (alt.internet.search-engines)
  • Re: Combining many records into 1
    ... NoteID is the same for each complete blog record ... SeqNum ranges from 1 - 20 for each NoteDetailID ... SQL Server MVP Anith Sen as a couple of methods on ... Books Online for SQL Server 2005 at ...
    (comp.databases.ms-sqlserver)
  • [ANN] Rassmalog 11.1.0
    ... It transforms blog entries written in Textile ... Josef ‘Jupp’ Schugt suggested making multi-word tags ... The feed method now allows you to create feeds containing ... only a single blog entry: ...
    (comp.lang.ruby)
  • Re: Session Variables - Good or Bad
    ... >> on a centralized model (no more so than any other SQL Server database). ... >> ISerializableAttribute() to the class) you simply change your web.config ... >> SerializableAttribute and an ISerializable interface. ... >> You can reach me thru my blog at ...
    (microsoft.public.dotnet.framework.aspnet)