I'm too young to say if it improved since brooks published his first book but I can say that the processes around project management improved.
At least in major corporations there is typically a huge intellectual capital (somewhat I hate that word...) involved in project management. Processes, documentation, evaluation.
When brooks wrote his book these weren't as good as they're now and they didn't have the experience.
Although with increased complexity in today's projects and hypes about various technologies and pressure on project managers by senior management I'm not sure if quality really imporved.
IMHO it's not important if quality increased but that we learn from our past mistakes and at least mitigate risks that we've already experienced. This way it can only become better.
Your cluelessness depends on your insights into the program you're submitting the bug for and not your education or other things you did in your life or list on your CV.
It's all not about reading the source. That's completely unimportant for just reporting the bug. But you have to know the program. There are too many bug reports filed by people who could be solved by reading the fine manual. (as they say)
To be honest with you, I don't care if Linux ever makes it on the desktop. What I care is that software works for me. It's not interesting if it was written by Microsoft, Apple, or some random geek who just happened to learn C yesterday. (Which doesn't mean that there aren't people at Microsoft or Apple that just happened to learn C yesterday...)
In another one of your posts you say that your time is valueable. You wouldn't guess it but other people's time is also valueable and if you want to get a bug fixed you should try to write good bug reports that are easy to understand and don't require multiple emails to be sent back and forth.
If you don't want to "know" the program and just want it to work for you then please don't contact the developer but submit a bug report as vague as you like to your distributor's bug tracking system and, depending on the distributor, they'll be happy to help you for a fee or even for free.
It all comes down to this: If people are not paid and do stuff in their free time and you want something from them, then you have to be polite and don't give them the impression of wasting their time.
There are many alpha versions of GNU projects out there packaged in various distributions although they're only served from alpha.gnu.org and ftp://alpha.gnu.org/README clearly states that it is not meant to be included in distributions.
I don't agree with your statement that "bug reports to distributors are often dead ends" but if that is really true you'll have to reproduce your problem first with the latest development version and not just submit a bug report that's for a version the developer released ages ago but happened to be packaged.
Keep in mind that it really takes away much of a developer's time to answer all these bug reports. (and most of them are not as useful as they could be)
IMHO bug reports to developers should only written by people who know what they are doing. (Compiled it themselves and really _know_ this program)
Please notice that users should use the distributor's bug tracking system.
Only people who are not clueless will use the tarballs or CVS exports from the developers and should submit directly to the developer's bug tracking system.
This way it all gets filtered at the distribution level because that's where the user got his program from.
This totally depends on where you got your version from. - If you got it directly from the developer by downloading sourceballs or CVS export you'll report it directly to the developer's bug tracking system.
If you use the distributor's version you'll simply report it to the distributor and he'll decide what to do with it.
This seems like a very good idea to me but it has one big problem. Like for every service provider you need equipment like NASes and multiple servers. I don't know how this should be financed for a non-profit ASP.
If it is possible to finance that it seems like a pretty interesting idea.
By reading this article I don't really get the difference between this search method used by gnutella and the harvest web indexer. I have to admit that I don't know much about both of them but for me it looks nearly the same.
Re:What is a "used" ebook?
on
RMS On eBooks
·
· Score: 1
I can think about one thing that has a much longer lifespan: paper.
Plain old paper is one of the best ways to store information if you don't have the time to make holes into stones.
Re:What is a "used" ebook?
on
RMS On eBooks
·
· Score: 1
Read again - I never said EBooks would be Free - only that they would be incredibly cheap.
As always: speech not beer.
What? That they need restitution to cover publishing and distribution costs??? Who is going to pay for them otherwise? The tooth fairy??
I even _want_ to pay for the distribution, but I think that the work should be free. It should be possible to order a book and share it with all the others in the local User Group or school and this is not possible with EBooks.
Well, you can view video or photographic images of most of the importantevents of the 20th century. There are at least one million hours of recorded newscasts. There is at least a petabyte of stored journalism. Run this by me again?????
There was a report a few months ago that (I think NYT) which stated that large archives of videos were lost. (If you store something on an audio or video tape it will not be there for long if you don't re-record it onto another tape every few years. Same for CDs, CDRs,... Paper seems to be one of the best choices.)
Re:What is a "used" ebook?
on
RMS On eBooks
·
· Score: 1
Think again.
1. EBooks are _NOT_/free/. You are not allowed to modify or update them. This is a major problem with technical publications.
2. EBooks cannot be borrowed by a University Library or some other library because they are only available electronically and will/disappear/ as soon as they become obsolete. (It's often really useful to get an obsolete book at library because humanity should learn from it's own faults. This wouldn't be possible with EBooks because they are destroyed)
3. Our century is already the worst documented one ever. (Heard that on TV last year). They stated that information will disappear with time because it is not archived on paper or anything else which lives longer than a few decades. I think we should do something against that.
Re:RMS should stick to coding
on
RMS On eBooks
·
· Score: 1
The fact of the matter is that copyrights are essential - for without them there would be no ownership of any creative works by the author. RMS's own GPL depends on the existance of copyright law. If there would be no copyright, rms wouldn't need a GPL.
If there would be no copyright law rms wouldn't need a GNU General Public License. We wouldn't need this license if there would be no copyright law. I think this would make some of his dreams true:-)
About a year ago there were Alpha 164sx computers on the market which had a large computational power but not that much of 2nd level cache and not high I/O performance. The price was the same as a good personal computer.
I think today you can get cheap alphas of this type on ebay or in the respective *.alpha newsgroups.
Are we slaves of Intel and Microsoft as Woz says in the Interview?
I don't think so. You have many Unix operating systems which run on a many hardware platforms. If you only think about Linux it is running on Intel, Alpha, StrongARM, Mac, UltraSparc, Sparc, and more. We don't want to make a list of the NetBSD ports here, right?:-)
I really think this interview was good, but this point was something which I didn't like. The rest of the interview was good work with really interesting answers. -thanks for it.
That's why you should always add 1900 to this year format:-)))
I think we will have a great time hacking all our perl scripts to handle these things right.
Seems I'm lucky. All my scripts seem to work right at the moment. (Also the shell scripts which use `date')
Not for OpenSource community
on
Under The Radar
·
· Score: 1
I already read this book a while ago and it seems that this was not written for the Open Source community or newbies to it.
I think O'Reilly's OpenSources is a _much_ better book because it contains the _opinions_ of the open source people and doesn't focus on redhat's history.
It's not bad, but the audience is a completely different. - It was interesting to read and I learned much more about Red Hat, but nothing I really need or want to know;-)))
This question is easy to answer. Debian is/was a GNU project.
In GNU's Bulletin, vol. 1 no. 21 you can read the following:
FSF and Debian Separate Amicably
Ian Murdock started the effort to put together Debian, a Linux-based GNU system designed to be easy to install & upgrade. He asked for & got the FSF's sponsorship for the project, hoping that besides being useful in its own right, it would give the FSF experience in packaging up a complete GNU system.
This March, Murdock stepped down as the head of Debian, having become too busy with other work. The new team head did not want FSF sponsorship. As a result, the FSF is no longer a sponsor of Debian. We wish the situation were otherwise. However, we are working together on some design issues.
The problem with GNU Hurd is that there are not enough people working on it. I myself tried to invest some time into it. I simply don't have enough free-time (if you can call it that way).
There is nobody working on GNU Hurd full-time. That's why it is so slow. But marcus invests _much_ time into it since 1996 (if I remember correctly) and is doing a _great_ job.
This was already a topic on the debian mailing lists. You are right. This makes it much easier for some people to install programs.
That's why the model of splitting the distribution from the applications is a good idea.
Look at FreeBSD's port collection. This is _not_ the best solution, but it is a beginning. It may be an option to create a core debian and the application-debian around it. This way people will have a stable distribution released every few months (the developers of this stable distro don't have to think about the applications). The applications will be available somewhere else. (That's possible with apt-get)
I just think that it's not good for debian as a whole to be that big. There are just too many bugs which are _not_ related to the core distribution.
There were many threads about that on debian-devel and I don't want to discuss that here. It was only an idea for a question.
Debian seems to be getting too big to be managed. (by the user and wakkerman)
There are many packages and they are getting more and more. ("What? There is a new window manager? - Package it!") - I don't think this is the responsibility of a distribution.
A distribution should be the base system to run linux. Every more advanced system should be installed by the unix administrator.
To make it even more worse, packages like netstd get split up in many others and packages which should be split, don't. (Look at tetex-bin. You only need xlib6g because there is xdvi in it. - If you drop xdvi in it's own package you don't have to install xlib6g and xbase on your servers)
Slashdot Mirror
I'd really like to know *why* they're doing that NOW. - I hope it's not because they're running out of money.
What would we do without an FSF?
I'm too young to say if it improved since brooks published his first book but I can say that the processes around project management improved.
At least in major corporations there is typically a huge intellectual capital (somewhat I hate that word...) involved in project management. Processes, documentation, evaluation.
When brooks wrote his book these weren't as good as they're now and they didn't have the experience.
Although with increased complexity in today's projects and hypes about various technologies and pressure on project managers by senior management I'm not sure if quality really imporved.
IMHO it's not important if quality increased but that we learn from our past mistakes and at least mitigate risks that we've already experienced. This way it can only become better.
Your cluelessness depends on your insights into the program you're submitting the bug for and not your education or other things you did in your life or list on your CV.
It's all not about reading the source. That's completely unimportant for just reporting the bug. But you have to know the program. There are too many bug reports filed by people who could be solved by reading the fine manual. (as they say)
To be honest with you, I don't care if Linux ever makes it on the desktop. What I care is that software works for me. It's not interesting if it was written by Microsoft, Apple, or some random geek who just happened to learn C yesterday. (Which doesn't mean that there aren't people at Microsoft or Apple that just happened to learn C yesterday...)
In another one of your posts you say that your time is valueable. You wouldn't guess it but other people's time is also valueable and if you want to get a bug fixed you should try to write good bug reports that are easy to understand and don't require multiple emails to be sent back and forth.
If you don't want to "know" the program and just want it to work for you then please don't contact the developer but submit a bug report as vague as you like to your distributor's bug tracking system and, depending on the distributor, they'll be happy to help you for a fee or even for free.
It all comes down to this: If people are not paid and do stuff in their free time and you want something from them, then you have to be polite and don't give them the impression of wasting their time.
There is no centralized bug reporting because the open source and free software communities are by design distributed.
You'll have to send bug reports to the entity you got your program from.
That's also the case with other packages.
There are many alpha versions of GNU projects out there packaged in various distributions although they're only served from alpha.gnu.org and ftp://alpha.gnu.org/README clearly states that it is not meant to be included in distributions.
I don't agree with your statement that "bug reports to distributors are often dead ends" but if that is really true you'll have to reproduce your problem first with the latest development version and not just submit a bug report that's for a version the developer released ages ago but happened to be packaged.
Keep in mind that it really takes away much of a developer's time to answer all these bug reports. (and most of them are not as useful as they could be)
IMHO bug reports to developers should only written by people who know what they are doing. (Compiled it themselves and really _know_ this program)
That's the best way to get ignored.
:) - and it is.
Writing a capital "URGENT" in the subject to a developer gets your mail dumped immediately.
Looks too much like spam
Please notice that users should use the distributor's bug tracking system.
Only people who are not clueless will use the tarballs or CVS exports from the developers and should submit directly to the developer's bug tracking system.
This way it all gets filtered at the distribution level because that's where the user got his program from.
This totally depends on where you got your version from. - If you got it directly from the developer by downloading sourceballs or CVS export you'll report it directly to the developer's bug tracking system.
If you use the distributor's version you'll simply report it to the distributor and he'll decide what to do with it.
It's all that easy.
This seems like a very good idea to me but it has one big problem. Like for every service provider you need equipment like NASes and multiple servers. I don't know how this should be financed for a non-profit ASP.
If it is possible to finance that it seems like a pretty interesting idea.
There is a generic mail interface called /usr/bin/mail. pls refer to `man mail'.
By reading this article I don't really get the difference between this search method used by gnutella and the harvest web indexer. I have to admit that I don't know much about both of them but for me it looks nearly the same.
I can think about one thing that has a much longer lifespan: paper.
Plain old paper is one of the best ways to store information if you don't have the time to make holes into stones.
As always: speech not beer.
What? That they need restitution to cover publishing and distribution costs??? Who is going to pay for them otherwise? The tooth fairy??
I even _want_ to pay for the distribution, but I think that the work should be free. It should be possible to order a book and share it with all the others in the local User Group or school and this is not possible with EBooks.
Well, you can view video or photographic images of most of the importantevents of the 20th century. There are at least one million hours of recorded newscasts. There is at least a petabyte of stored journalism. Run this by me again?????
There was a report a few months ago that (I think NYT) which stated that large archives of videos were lost. (If you store something on an audio or video tape it will not be there for long if you don't re-record it onto another tape every few years. Same for CDs, CDRs, ... Paper seems to be one of the best choices.)
Think again.
/free/. You are not allowed to modify or update them. This is a major problem with technical publications.
/disappear/ as soon as they become obsolete. (It's often really useful to get an obsolete book at library because humanity should learn from it's own faults. This wouldn't be possible with EBooks because they are destroyed)
1. EBooks are _NOT_
2. EBooks cannot be borrowed by a University Library or some other library because they are only available electronically and will
3. Our century is already the worst documented one ever. (Heard that on TV last year). They stated that information will disappear with time because it is not archived on paper or anything else which lives longer than a few decades. I think we should do something against that.
If there would be no copyright law rms wouldn't need a GNU General Public License. We wouldn't need this license if there would be no copyright law. I think this would make some of his dreams true :-)
About a year ago there were Alpha 164sx computers on the market which had a large computational power but not that much of 2nd level cache and not high I/O performance. The price was the same as a good personal computer.
I think today you can get cheap alphas of this type on ebay or in the respective *.alpha newsgroups.
Is there really a "Laptop System" install in redhat-6.2? I haven't seen one in the redhat-6.2de beta version.
Are we slaves of Intel and Microsoft as Woz says in the Interview?
:-)
I don't think so. You have many Unix operating systems which run on a many hardware platforms. If you only think about Linux it is running on Intel, Alpha, StrongARM, Mac, UltraSparc, Sparc, and more. We don't want to make a list of the NetBSD ports here, right?
I really think this interview was good, but this point was something which I didn't like. The rest of the interview was good work with really interesting answers. -thanks for it.
That's why you should always add 1900 to this year format :-)))
I think we will have a great time hacking all our perl scripts to handle these things right.
Seems I'm lucky. All my scripts seem to work right at the moment. (Also the shell scripts which use `date')
I already read this book a while ago and it seems that this was not written for the Open Source community or newbies to it.
;-)))
I think O'Reilly's OpenSources is a _much_ better book because it contains the _opinions_ of the open source people and doesn't focus on redhat's history.
It's not bad, but the audience is a completely different. - It was interesting to read and I learned much more about Red Hat, but nothing I really need or want to know
This question is easy to answer. Debian is/was a GNU project.
In GNU's Bulletin, vol. 1 no. 21 you can read the following:
FSF and Debian Separate Amicably
Ian Murdock started the effort to put together Debian, a Linux-based GNU system designed to be easy to install & upgrade. He asked for & got the FSF's sponsorship for the project, hoping that besides being useful in its own right, it would give the FSF experience in packaging up a complete GNU system.
This March, Murdock stepped down as the head of Debian, having become too busy with other work. The new team head did not want FSF sponsorship. As a result, the FSF is no longer a sponsor of Debian. We wish the situation were otherwise. However, we are working together on some design issues.
The problem with GNU Hurd is that there are not enough people working on it. I myself tried to invest some time into it. I simply don't have enough free-time (if you can call it that way).
There is nobody working on GNU Hurd full-time. That's why it is so slow. But marcus invests _much_ time into it since 1996 (if I remember correctly) and is doing a _great_ job.
This was already a topic on the debian mailing lists. You are right. This makes it much easier for some people to install programs.
That's why the model of splitting the distribution from the applications is a good idea.
Look at FreeBSD's port collection. This is _not_ the best solution, but it is a beginning. It may be an option to create a core debian and the application-debian around it. This way people will have a stable distribution released every few months (the developers of this stable distro don't have to think about the applications). The applications will be available somewhere else. (That's possible with apt-get)
I just think that it's not good for debian as a whole to be that big. There are just too many bugs which are _not_ related to the core distribution.
There were many threads about that on debian-devel and I don't want to discuss that here. It was only an idea for a question.
Debian seems to be getting too big to be managed. (by the user and wakkerman)
There are many packages and they are getting more and more. ("What? There is a new window manager? - Package it!") - I don't think this is the responsibility of a distribution.
A distribution should be the base system to run linux. Every more advanced system should be installed by the unix administrator.
To make it even more worse, packages like netstd get split up in many others and packages which should be split, don't. (Look at tetex-bin. You only need xlib6g because there is xdvi in it. - If you drop xdvi in it's own package you don't have to install xlib6g and xbase on your servers)