Announcement

Collapse
No announcement yet.

Exactly why do we need mark up for data distribution?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • harmor
    replied
    You can take a horse to water but you can't make it drink.

    Leave a comment:


  • Wayne Luke
    replied
    Then you are lucky they provided a format that you can use. Even with that, I can import the XML into Excel, Word, or any other application and format it as a table quickly and easily. Try importing your CSV into Word and formatting it easily.

    Obviously you care about bandwidth and the CSV fits your needs so that is what you are going to use. Most people don't care what format the data is delivered in as long as they can view it and they certainly don't care how much bandwidth it is taking up from the server.

    Leave a comment:


  • Mike Sullivan
    replied
    Originally posted by Zecherieh
    This is a perfect example - since they have both

    http://www.repole.com/sun4cast/stats/ncaa2005lines.xml = 136K

    http://www.repole.com/sun4cast/stats/ncaa2005lines.csv = 29K

    Same exact Data
    To be technical, this is not the exact same data. Data stored by the XML that the CSV doesn't store:
    • League Name
    • Year (possible to infer in this case, though not so much if it's a partial set and a season that spans the year end)
    • Season Type
    • Team Number (useless, but they included it)
    • Additional game info ("@ Reliant Stadium")


    Trying to fit all of those pieces of data into the same CSV file would not work (the first 3 at least). Also worth noting that this XML file has more markup than is necessary.

    Regarding your original question, XML is much more suited to complex data, as mentioned. CSV cannot easily represent multiple data formats in one file. Especially when dealing with 1:N relationships, which XML excels at. You could store the an entire forum-thread-post tree in an XML file without relying on an ID number.

    Compared to a serialized array, it's much easier to edit and read anyway. Editing serialized stuff manually is not fun, and reading it isn't either. From a parsing perspective, XML can be parsed without slurping the entire thing into memory if desired.

    Leave a comment:


  • Zecherieh
    replied
    Originally posted by Wayne Luke
    For this specific data feed, yes.

    If I was looking for a quick line to place a bet on upcoming games, yes it is useful. I don't need to store and manipulate the data because it will be obsolete next week when the teams play different games.

    That is what newsfeeds are for: Quick up to the minute information that can be discarded as soon as it is used. Long term storage and formatting of newsfeeds is not needed. Once it scrolls off the screen it isn't needed anymore.
    Except one thing - in my world - that data is much more useful now - then it was the week of the game (that data is the entire NCAA Season by the way)

    And that data is not even going to be available from that source until the week is complete and in the books - and you can not place a bet on that line. (yes - I live in a strange world.)

    Leave a comment:


  • Wayne Luke
    replied
    Originally posted by Zecherieh
    .. but is it really useful - in either format - this way ?

    Not to me. I could not see any user being able to utilize, or even having a desire to utilize this data this way. Its just a delivery method - for us then to re organize, reformat - and display in a way that our users can use it and find it useful.
    For this specific data feed, yes.

    If I was looking for a quick line to place a bet on upcoming games, yes it is useful. I don't need to store and manipulate the data because it will be obsolete next week when the teams play different games.

    That is what newsfeeds are for: Quick up to the minute information that can be discarded as soon as it is used. Long term storage and formatting of newsfeeds is not needed. Once it scrolls off the screen it isn't needed anymore.

    Leave a comment:


  • Zecherieh
    replied
    Originally posted by Wayne Luke
    And I can read the first in my brower and the second requires yet another application.

    The first I can load, press CTRL-F and find what I want. No problem.

    The second, I have to load into Excel, format so it is readable, then search for what I want. Then what happens if the person doesn't have a spreadsheet program?
    .. but is it really useful - in either format - this way ?

    Not to me. I could not see any user being able to utilize, or even having a desire to utilize this data this way. Its just a delivery method - for us then to re organize, reformat - and display in a way that our users can use it and find it useful.

    Leave a comment:


  • DirectPixel
    replied
    Originally posted by Zecherieh
    I am sure that if I was more of a Visual Studio type programmer - I would love XML - Microsoft had made XML easy to use years ago - but I am not

    More of a PHP kind of guy - and we aint had it so easy over the years - and now that its supposed to be easier - I am probably to stuck in my ways
    Yep, PHP's got some nice XML support built-in these days.

    Leave a comment:


  • Zecherieh
    replied
    I am sure that if I was more of a Visual Studio type programmer - I would love XML - Microsoft had made XML easy to use years ago - but I am not

    More of a PHP kind of guy - and we aint had it so easy over the years - and now that its supposed to be easier - I am probably to stuck in my ways

    Leave a comment:


  • Zecherieh
    replied
    Originally posted by DirectPixel
    Imagine if the number of fields were increased from 3 to 30, with the other 20 of those fields being numeric. And you don't have Excel installed.
    I aint gonna find anything useful with this sort of data in either regards - even now with it in Excel - its rather useless - as it is - I would just throw it into a database ...

    ... which when using MySQL - the code to get into the database for the CSV is going to 75 percent the connection code - then the one line to import the csv...

    Leave a comment:


  • Zecherieh
    replied
    Originally posted by Wayne Luke
    Not really. Just need to change the client to a listener and have the server publish its results on an occasional basis. When it republishes, it sends a notice to the client to retrieve the new data.
    In that regards - to even avoid the listening .... this is strictly for content delivery from server to server

    The party with the data can just send a ping, or a connection to a specified file on the receiver's server - which then initiates the receiver to reach out and grab the format (or not if its not something the receiver wants) - that it needs - eliminating the constant checking that goes on by the receiver -

    and actually allows the receiver to get the data quicker - as they know as soon as its available - not just if its available when they check for new data every five minutes, ten minutes, 30 seconds, whatever.
    Last edited by Zecherieh; Thu 9 Feb '06, 7:05am.

    Leave a comment:


  • DirectPixel
    replied
    Originally posted by Zecherieh
    This is a perfect example - since they have both

    http://www.repole.com/sun4cast/stats/ncaa2005lines.xml = 136K

    http://www.repole.com/sun4cast/stats/ncaa2005lines.csv = 29K

    Same exact Data
    Imagine if the number of fields were increased from 3 to 30, with the other 20 of those fields being numeric. And you don't have Excel installed.

    Leave a comment:


  • Wayne Luke
    replied
    Originally posted by Zecherieh
    This is a perfect example - since they have both

    http://www.repole.com/sun4cast/stats/ncaa2005lines.xml = 136K

    http://www.repole.com/sun4cast/stats/ncaa2005lines.csv = 29K

    Same exact Data
    And I can read the first in my brower and the second requires yet another application.

    The first I can load, press CTRL-F and find what I want. No problem.

    The second, I have to load into Excel, format so it is readable, then search for what I want. Then what happens if the person doesn't have a spreadsheet program?

    Leave a comment:


  • Zecherieh
    replied
    (and the really ironic part is - since excel takes over on the CSV - the CSV data is way more readable than the XML data )

    Leave a comment:


  • Wayne Luke
    replied
    Originally posted by DirectPixel
    Changing XML delivery from a user-get to a server-push medium is quite a task.
    Not really. Just need to change the client to a listener and have the server publish its results on an occasional basis. When it republishes, it sends a notice to the client to retrieve the new data.

    Leave a comment:


  • Zecherieh
    replied
    This is a perfect example - since they have both

    http://www.repole.com/sun4cast/stats/ncaa2005lines.xml = 136K

    http://www.repole.com/sun4cast/stats/ncaa2005lines.csv = 29K

    Same exact Data

    Leave a comment:

widgetinstance 262 (Related Topics) skipped due to lack of content & hide_module_if_empty option.
Working...
X