Charlotte Foust
cfoust at infostatsystems.com
Thu Sep 3 13:33:48 CDT 2009
On the other hand, why would you even think of using only one file? We write tables or table changes to xml files for each table and read them in appropriately at the other end. Could it be you're caught in the mdb mode of putting your eggs all in one basket? Charlotte Foust -----Original Message----- From: accessd-bounces at databaseadvisors.com [mailto:accessd-bounces at databaseadvisors.com] On Behalf Of jwcolby Sent: Thursday, September 03, 2009 10:12 AM To: Access Developers discussion and problem solving Subject: Re: [AccessD] XML >However, for transferring data, storing things like individual settings, etc., it is quick, easy and efficient. I will give you the quick and easy part, particularly for the individual settings part. In terms of transferring data, it is just a bumbling mass of inefficiencies. Try sending a hundred million records of 640 fields and see if you can even fit the file on a terabyte disk. The answer will be NO! XML is rapidly being adopted for transferring data across the internet and yet it expands the size of the data by a factor of 10, or even 100. It is simply ludicrous for such uses. Pipe delimited CSV files are an order of magnitude more efficient for that kind of thing. Now I will grant you that you do not get the data type etc with that but still... John W. Colby www.ColbyConsulting.com Charlotte Foust wrote: > And I wasn't talking about storing a whole database in it. However, > for transferring data, storing things like individual settings, etc., > it is quick, easy and efficient. It is CERTAINLY as efficient as > delimited text for such things since the adapters are already built > into .net, so there's no need to create them one off for each effort. > You don't have to read the xml into a particular structure because the > structure is built in. And we store SQL in xml files where it can be > quickly retrieved by our data tier but is compiled into the dll > putting it beyond the reach of exploring users. > > Charlotte Foust > > -----Original Message----- > From: accessd-bounces at databaseadvisors.com > [mailto:accessd-bounces at databaseadvisors.com] On Behalf Of jwcolby > Sent: Thursday, September 03, 2009 6:06 AM > To: Access Developers discussion and problem solving > Subject: Re: [AccessD] XML > > I have to say I agree with Stuart on this one. XML is about as > inefficient as you can get for storing data. > > We as database designers strive to select the smallest possible > datatype to hold whatever data we are storing in the data store. Now > you export a long integer to XML and it turns into hundreds of bytes. > And all so it can be man readable? When was the last time you > actually read an XML document? What percentage of all XML documents > do you (or any human) ever actually read (in XML format)? > > It is almost as if the hard disk consortium got together in a secret > room deep in a mountain in Russia, surrounded by KGB security > specially hired to keep their meeting private... and designed a "storage system" > to help them stimulate sales. > > "Hard drives have gotten so big that people are only buying one. > Let's design this system called XML that will take anything and store > it in layers of wrappers that will expand the original size by 1000. > We will sell many more disks now..." > > "Now let's leak it's existence to MS and tell them that it has already > become the next storage standard..." > > "Ahh... our plan worked, MS is now storing the world in XML." > > Disk drive manufacturer stock prices skyrocket, approaching the share > price of Berkshire Hathaway.. > > High fives around... > > John W. Colby > www.ColbyConsulting.com > > -- AccessD mailing list AccessD at databaseadvisors.com http://databaseadvisors.com/mailman/listinfo/accessd Website: http://www.databaseadvisors.com