I made the mistake of going with Bell Atlantic for my DSL service, and I think it ranks as the #1 worst customer service experience of my life.
To give one example, I recently had an outage which lasted a week. I clocked over 11 hours on the phone with Bell Atlantic during that week trying to get the matter resolved. I've had six such outages in the space of around four months.
Warning for Linux users: Bell Atlantic's sales staff will tell you "It's OK to use Linux, but you're responsible for the configuration at your end." Tech support, however, says "We don't support Linux" and will adamantly refuse to help you, even if you can clearly demonstrate that the problem is at their end. I've had to raise a dreadful row every time.
It seems to me that there's more than one way of speculating here:
1. You can simply extrapolate from current trends, as most of the posts in this thread so far have done.
2. You can recognize we're not able to answer the question in our current state of knowledge, but that we can at least say what kind of answer is needed.
Imagine that each computer is a single air molecule. Nearly all of our work so far has been on the behavior of individual air molecules, with some work on interaction between perhaps two of them. However, in a system this big and complex, there are likely to be large-scale behaviors analogous to sound waves, which we haven't predicted or identified.
We'll have to wait and see what sorts of behaviors these will be, and what practical impact they will have; we've only got a few initial observations, because the air is new.
This doesn't directly answer your question, but a relevant comment here.
I've thought for a while that it would be a good idea if there were were some sort of graphical programming: you drag together some sort of graphical blocks to represent loops, procedures, typed variables, etc. rather than entering text. The program compiles to ordinary machine code, but the representation at the programmer's end is something completely different from text.
I don't know whether something like this has been tried, but I expect that it would be a gentler way to develop the cognitive skills of programming without forcing the learner to grapple with so many technical details at first. I'm sure that children would not be the only ones who could benefit.
A suggestion: contact Bank of America and tell them how you feel about their legal threats against the creator of the Dialectizer. Here is how to reach them:
1. Go to http://www.bankofamerica.com/ 2. Click on "Contact Us", and choose your state. 3. Go down to the bottom of the second column and click on "General Inquiries and Customer Service". 4. At the top of the page, click on "Click here to send a secure e-mail using SSL to Bank of America". 5. Write your message, using "Comments and Suggestions" for your subject.
So if I'm understanding, you don't see any problem if we all go on indefinitely installing two separate largish libraries on our systems, even tho the two libraries have essentially the same functionality. I see this as a waste of disk space and virtual memory. In that sense, yes, duplication is bad.
But let's take the view of many commercial developers that disk space and memory keep getting cheaper, so there's no real need to conserve these resources. In that case, I fall back on my argument that there is a finite amount of talent and effort in the free software community, and developing two large libraries with essentially the same functionality is a waste of time and talent.
Personally, I don't see how the "strength in diversity" reasoning applies to the KDE/Gnome divide.
Is it a good thing that I, as an application developer, have to code, debug, and maintain my application for what amounts to two separate platforms, when I'm really trying to target "Linux"?
I don't understand why it's a good thing that two teams of programmers are implementing essentially the same functionality (with small differences), when they could be working together and getting twice as far in the same amount of time.
To what extent is support for Unicode being built into Gnome? What long-term plans are there in this area? As someone who frequently has need to mix multiple scripts in a single document, I am still looking forward to a platform which makes this task convenient.
I'd love to work full-time toward better Unicode support in Gnome, but right now I can only take on smallish tasks. Is there a list somewhere of tasks toward this goal which individuals with an hour or two to spare now and then could work on?
Some of the folks who responded earlier that "Coke" is not a trademark because it is just a nickname that consumers came up with. Let me say first that I'm not an attorney. Also, everything I'm about to say applies in the U.S.; I don't know about trademark law in Switzerland. What I'm about to say is based on a book that I read about trademark law a few years ago, so some of the law may have evolved since then. In any case, the use of the term "Coke" for "Coca-Cola" is put forward in the books as a textbook example of a case where a company has rights to a trademark without having created the name itself. Since the term "Coke" is popularly understood to refer to Coca-Cola, the Coca-Cola Company could have successfully sued another company for marketing a beverage called "Coke" (The Coca-Cola company did in fact later file for a trademark on "Coke" to reinforce its position). The basic question in trademark law is whether someone is unfairly benefiting from someone else's product image. As for totally different products having the same name, the question here is more subtle. If your name draws on general cultural motifs (say, the names of Greek gods) or general positive words such as "Imperial", you can't do anything if someone else markets a totally different product with the same name (so if I run Imperial Cleaners, I can't sue you for trademark infringement if you sell Imperial Tomatoes). However, if the word is one you coined yourself, it's different. Someone once tried to market Kodak brand cigarette lighters, and the Kodak-Eastman Company successfully sued because it was a term it came up with itself. Under U.S. law, it looks to me like a site called "coke.com" which is about cocaine is not in violation of Coca-Cola's trademark, because "coke" is widely understood as a term for cocaine, and the individual in question can't reasonably be said to be benefitting unfairly from Coca-Cola's corporate image. I can't say whether the same is true under Swiss law.
Slashdot Mirror
To give one example, I recently had an outage which lasted a week. I clocked over 11 hours on the phone with Bell Atlantic during that week trying to get the matter resolved. I've had six such outages in the space of around four months.
Warning for Linux users: Bell Atlantic's sales staff will tell you "It's OK to use Linux, but you're responsible for the configuration at your end." Tech support, however, says "We don't support Linux" and will adamantly refuse to help you, even if you can clearly demonstrate that the problem is at their end. I've had to raise a dreadful row every time.
1. You can simply extrapolate from current trends, as most of the posts in this thread so far have done.
2. You can recognize we're not able to answer the question in our current state of knowledge, but that we can at least say what kind of answer is needed.
Imagine that each computer is a single air molecule. Nearly all of our work so far has been on the behavior of individual air molecules, with some work on interaction between perhaps two of them. However, in a system this big and complex, there are likely to be large-scale behaviors analogous to sound waves, which we haven't predicted or identified.
We'll have to wait and see what sorts of behaviors these will be, and what practical impact they will have; we've only got a few initial observations, because the air is new.
I've thought for a while that it would be a good idea if there were were some sort of graphical programming: you drag together some sort of graphical blocks to represent loops, procedures, typed variables, etc. rather than entering text. The program compiles to ordinary machine code, but the representation at the programmer's end is something completely different from text.
I don't know whether something like this has been tried, but I expect that it would be a gentler way to develop the cognitive skills of programming without forcing the learner to grapple with so many technical details at first. I'm sure that children would not be the only ones who could benefit.
1. Go to http://www.bankofamerica.com/
2. Click on "Contact Us", and choose your state.
3. Go down to the bottom of the second column and click on "General Inquiries and Customer Service".
4. At the top of the page, click on "Click here to send a secure e-mail using SSL to Bank of America".
5. Write your message, using "Comments and Suggestions" for your subject.
I'd be respectful but firm.
But let's take the view of many commercial developers that disk space and memory keep getting cheaper, so there's no real need to conserve these resources. In that case, I fall back on my argument that there is a finite amount of talent and effort in the free software community, and developing two large libraries with essentially the same functionality is a waste of time and talent.
Is it a good thing that I, as an application developer, have to code, debug, and maintain my application for what amounts to two separate platforms, when I'm really trying to target "Linux"?
I don't understand why it's a good thing that two teams of programmers are implementing essentially the same functionality (with small differences), when they could be working together and getting twice as far in the same amount of time.
I'd love to work full-time toward better Unicode support in Gnome, but right now I can only take on smallish tasks. Is there a list somewhere of tasks toward this goal which individuals with an hour or two to spare now and then could work on?
Some of the folks who responded earlier that "Coke" is not a trademark because it is just a nickname that consumers came up with. Let me say first that I'm not an attorney. Also, everything I'm about to say applies in the U.S.; I don't know about trademark law in Switzerland. What I'm about to say is based on a book that I read about trademark law a few years ago, so some of the law may have evolved since then. In any case, the use of the term "Coke" for "Coca-Cola" is put forward in the books as a textbook example of a case where a company has rights to a trademark without having created the name itself. Since the term "Coke" is popularly understood to refer to Coca-Cola, the Coca-Cola Company could have successfully sued another company for marketing a beverage called "Coke" (The Coca-Cola company did in fact later file for a trademark on "Coke" to reinforce its position). The basic question in trademark law is whether someone is unfairly benefiting from someone else's product image. As for totally different products having the same name, the question here is more subtle. If your name draws on general cultural motifs (say, the names of Greek gods) or general positive words such as "Imperial", you can't do anything if someone else markets a totally different product with the same name (so if I run Imperial Cleaners, I can't sue you for trademark infringement if you sell Imperial Tomatoes). However, if the word is one you coined yourself, it's different. Someone once tried to market Kodak brand cigarette lighters, and the Kodak-Eastman Company successfully sued because it was a term it came up with itself. Under U.S. law, it looks to me like a site called "coke.com" which is about cocaine is not in violation of Coca-Cola's trademark, because "coke" is widely understood as a term for cocaine, and the individual in question can't reasonably be said to be benefitting unfairly from Coca-Cola's corporate image. I can't say whether the same is true under Swiss law.