Hobbyists like to tinker with custom computing configurations and experiences, e.g., Linux desktop.  Most people don't, e.g., Linux desktop.

There will always be a niche for hobbyist computing experiences, aka the long tail, but there is a much larger market in fulfilling the mainstream computing experience.

Yahoo and Hotmail have more marketshare than Gmail, because both existed earlier.  Gmail also hurt their marketshare by making the service invite-only in the beginning.  Being invite-only however contributed to the "exclusive" status of gmail addresses.  That status persists even though anyone can sign up for a gmail email address nowadays.

Exclusivity also contributed to the quality ("non-spamness") of emails.  Since anyone could sign up for yahoo and hotmail addresses, spammers would, and, consequently, some email from yahoo and hotmail addresses would be from spammers.  This became a problem for gmail too once they dropped their exclusivity.

Jul 02, 2008 at 5:43 AM, Bas wrote

As stated above, this thread is centuries old.

Edit: Ah never mind, sorry. I saw your request about needing the solution to be deployable after posting.

var data =
    from item in list
    group item by item.Key into gruppe
    from item in gruppe
    orderby gruppe.Key, item.Value
    select item; 

Technically, you don't need to group them:

var data =
    from item in list
    orderby item.Key, item.Value
    select item;

5 days ago, spivonious wrote

@PerfectPhase: And just so you know, the "select item" part isn't needed.

That isn't correct.  I get a "query body must end with a select clause or a group clause" error without it.

As for bandwidth, wireless Bluetooth can handle HID fine so I'm sure wired USB can too.

Anyways, the UI latency you've noticed is likely caused by application-level latency, not hardware-level or driver-level latency.  It's easy to see this if you play a game with light system requirements.

Others: Use "msSiteModeCreateJumplist" in window.external instead.

IBM had a similar dilemma with mainframe applications, and even though mainframe applications are not sexy or innovative, they're still making money for IBM and others involved in the care and maintenance of such software.

20 years from now, Google may have the same dilemma with Web applications when the "next big thing" displaces them.