Linked by Thom Holwerda on Thu 10th Jan 2008 23:14 UTC
Thread beginning with comment 295285
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
*Individual applications are still off on their own not playing nice with other applications.
DBUS is doing quite well in this regard, though it is down to the application owners to be sympathetic to the user and accommodate DBUS.
*Users are still at the mercy of developers who may or may not have the insight to create their applications in a way that allows them to solve their specific problems.
Not sure what you mean here? All software is written to "solve a problem"... if a piece of software doesn't fit someone's need, at least with OSS the user can submit a bug report/feature request direct to the developer(s).
*Security is still a big issue and will continue to become more of an issue as programs get more and more complex.
Agreed, but recent OS' include complex systems to alleviate security risks... plus most have a secure foundation of user privilege restriction and threat limitation.
*A complete re-think of a lot of applications will be required to make use of the change to highly parallel computing.
Not sure this is necessary - most software relies on other libraries/components to do the hard work, and most libraries who have this dependency are already working on fully supporting multiple cores at least (look at the latest gstreamer codec packs for e.g.)
*Most businesses are still doing word processing as their major use of computers. This word processing is based on the idea of making paper documents which is a huge waste.
Eh? So writing reports to distribute, financial statements, press releases, marketing material, software documentation.... all sorts of things that need to be distributed on paper to people? This boils down to cost and technology - even amazon's electronic reader (and similar products) won't solve the use of paper!
I'm not sure your points really summarise the limitations of modern computing... merely the challenges which have mostly been addressed.
Not sure what you mean here? All software is written to "solve a problem"... if a piece of software doesn't fit someone's need, at least with OSS the user can submit a bug report/feature request direct to the developer(s).
I think he means that all software is written by programmers and not necessarily domain expertes. I guess the theory is that if programming was somehow 'easier' then architectula visulaization software for example could be written by architects who have detailed knowledge of what they do and don't need, and who could also tweak the program to their indivdual projects.
Instead the software has to be written by developers who may know a lot about 3d graphics, but very little about the needs and requirements of architects for various projects. So he has to try to talk to the architects and glean what they need, and the architects have to try to understand what is and isn't possible to do with the current level of technology. A process which is often slow, frustrating for all involved and fraught with the risk of misunderstanding. All of which leads to crap software for the architect to use.






Member since:
2005-08-17
Granted, I doubt it'd go on forever. Word 97/Excel 97/etc functionality is all most people need. Eventually, people will stop working on the OS-level and GUI-level (which will be "good enough" for 95% of people) and start working about the really interesting problems.
In the past 10 years the hardware has changed a lot. But the software and the way it interacts with the user hasn't changed much at all and still feels inefficient and annoying.
There is plenty to be done in fixing the fundamentals of the OS and the basics of how people interact with their computers.
*Individual applications are still off on their own not playing nice with other applications.
*Users are still at the mercy of developers who may or may not have the insight to create their applications in a way that allows them to solve their specific problems.
*Security is still a big issue and will continue to become more of an issue as programs get more and more complex.
*A complete re-think of a lot of applications will be required to make use of the change to highly parallel computing.
*Most businesses are still doing word processing as their major use of computers. This word processing is based on the idea of making paper documents which is a huge waste.
These are still fairly interesting problems that still need solving.