Also worth noting: in Windows you will eventually probably find it necessary to clean up all the junk you've downloaded, and you also have to search through all the junk in the first place to find your download.
Its hard to see how using Synaptic is more difficult than the Windows approach...
In the long term the savings in the total number of steps required and time to complete a task such as this, can be quite significant.
Methinks this is a case where a little bit of effort (e.g. in setting up your download repositories if needs be) has a substantial payoff!
Many features of a DBMS are highly advantageous from the point of view of an IMAP server. The obvious performance differential between the database options and both UW and Courier indicates that email storage is indeed a problem well suited for a database solution. Indexing capabilities give Cyrus and mySQL an advantage over Courier and UW when scanning headers and searching header fields. mySQL's full-text index provides a particularly expedient method for searching through message text, although it adds significant maintenance cost to operations such as adding and removing messages. A server-side buffer cache also improves performance by speeding up searches on recently accessed data. Although UW outperforms Cyrus by a small margin on some full-text searches, mySQL demonstrates clearly that a DBMS can search email much more quickly than a file-based solution. Most importantly, these results offer desperately needed empirical data comparing the performance of these three storage implementations.
Slashdot Mirror
Also worth noting: in Windows you will eventually probably find it necessary to clean up all the junk you've downloaded, and you also have to search through all the junk in the first place to find your download.
Its hard to see how using Synaptic is more difficult than the Windows approach...
In the long term the savings in the total number of steps required and time to complete a task such as this, can be quite significant.
Methinks this is a case where a little bit of effort (e.g. in setting up your download repositories if needs be) has a substantial payoff!
Yes. Especially when emacs needs to be revivified.
There is http://www.delorie.com/gnu/docs/emacs/viper_26.htm l/ for those wanting to return to the one true path (and avoidance of carpal tunnel syndrome).
A little bit of research reveals this:
http://www.usenix.org/events/lisa03/tech/full_pape rs/elprin/elprin_html/
I don't suppose anyone's come across any newer research (or implementations using this approach)?!?