Kawahee, you must not be a programmer. If you were, you would know that one of the key aspects of an effective programmer is laziness. Developers manifest this quality by spending inordinant hours researching (or less successfully asking/.), downloading and testing various tools built by other programmers in order to find the 'right thing'. If, and only if they can not find a tool that does a very high percentage of what they want, then they will break down and spend the time to build it themselves. More commonly, they will use the 'high percentage' tool, and build a smaller add-on to capture that last few percentage points.
This, of course, assumes the developer does not work in a shop where the choice of tools is dictated by management.
Under X windows (linux/unix - kde, gnome, motif etc...) -
1. Highlight text with left mouse button.
2. Go to new location and click middle mouse button to paste.
Voila! Your text is copied and pasted - probably the most efficient method available.
The DOS/Windows command line comes close:
1. Highlight text with left mouse button.
2. Right click in window to copy.
3. Right click in window to paste.
But this only works inside the CLI. If you want to take it out to another window, you are still stuck using a windows shortcut (CTRL V) or the right click menu 'paste' option.
With new complex applications running on platform architectures that are constantly changing (for such things as convergent applications), decentralizing your platform would be insane.
Your costs to change the system would go up. Your security would be exponentially more costly, and your telemetry and other time sensitive aspects of the management solution would go through the roof!
I would also hazzard a guess that removal of the datacenter would probably violate SOX (it would probably be a bad thing for Johnny in accounting to have physical access to the server housing his data).
This is the problem with OSS. Everyone wants to get famous for the next big breakthrough and nobody wants to maintain the shit.
Whoa partner! You're using an awful broad brush to paint your picture.
Debian, Slackware, Ubuntu and other distros are rock solid and continually maintained.
Emacs, Python, Firefox, Open Office are some very big applications that are well maintained.
So - while some OSS projects may have serious maintenance problems, you can't say that for all of them. Additionally, I wouldn't be so quick, as some have, to believe proprietary software is any better maintained; I would argue just the opposite.
The direction MS is going with windows will definitely not fill that niche - particularly for businesses that have to be cost concious and can't blow thousands or millions of dollars on new installations that support the latest and greatest from MS.
I wouldn't load Win98 on a 486 66 - even in 1998 - and as I mentioned that system only ran DOS before I got it. I do believe you can build a minimal Linux system on a Pentium 100 or better system that can appear just as full featured on the GUI as Win98 systems. The problem is no one in the Linux arena has that as a goal, that I am aware of (educate me if there are - as an aside, I do know there is a windows work-alike project called ReactOS that is going for binary equivalence to run win32 apps - and already has many native windows apps - MS Powerpoint, Adobe Photoshop etc.. - that will run on it).
To start such a project in Linux, you could list all the things you want to see on the system from an interface and tools perspective that a typical Win98 user would expect - and build to that standard. You would reach into your embedded systems bag of tricks to build highly tuned kernels that are fast, efficient, with a small footprint designed for those low-end processors. You would tweak the window manager to be more 'windows-like' (start with fvwm'95 or other lightweight window managers). On top of that you would add all the requirements for strange drivers - again the embedded approach would have merit - minimalistic and targeted. Finally, you would probably need to load WINE or Crossover Office in order to run ancient DOS and Windows apps that interface with the hardware or data produced by it - unless, of course, you could reverse engineer these applications and build them natively as a linux binary. I can see why this appears at first glance to be a daunting task.
Unfortunately, I haven't seen anything on the radar that targets this - which could be a missed opportunity for integrators - and a danger for those Win98 systems still attached to the network - as security erodes, or even those doing critical tasks off the net could be lost to corruption as drives fail over time - OEM installation disks notwithstanding. At some point these businesses will have to make a move - either replacing all the old gear, or spending the time building an open source solution that can have a lifespan longer than dictated by a vendor.
Not an easy problem; thankfully I don't have that problem to deal with - hence why I thought I could spend a few moments pointing out some alternatives.
Not even the smallest distros provide you for that (according to recomendations... however, once I have time, I'll try to get it to run with XFCE or some *box).
I hacked the zipslack Slackware distro to load on my Acer 486 66mhz laptop (20 MB ram, 500 MB hard drive); I have a parallel port iOmega 100 MB Zipdrive I used to hold the needed packages for a minimal system, and used the regular boot disk with the Parallel port root disk - to gain access to the drive.
Once I got the minimal system running, I transfered the packages I required from the distro via the zipdrive, and used installpkg to flesh out the applications I wanted from the regular distro.
As a finishing touch I built the TWIN text based based window manager - which is implimented in curses and emulates X11 interfaces - on my workstation, then transfered the working window manager over. X.org would not fit on the system given the limitations of space and video capability of that system. It runs great - much better than the DOS 4 that was single-tasking on it when I got it.
So it is possible to breath new life into very marginal (by today's standards) machines.
The big problem today is most highschools do not provide computer programming as part of their cirriculum. Instead they teach students to run MS Windows, 'Word', 'Excel', and browsers, maybe a graphics art class using photoshop, and if they are real lucky a computer hardware class where they learn to build a PC from parts, troubleshoot components and maybe build an electronic circuit from parts - none of which helps students become programmers. So they are on their own, for the most part, and enter CS degree program without a clue - if they enter it at all (CS major enrollments are down for a number of factors including offshoring - but largely due to lack of opportunities in CS in public schools).
Back in the 'good old days' - if you were lucky enough to have a microcomputer (the TI 99a my parents bought for me in 1981 was a whopping $2000!! - the better part of the value of a car back then) or access to microcomputers in a fledgling 'computer science' class (I was lucky enough to also attend the first CS class provided at my highschool) - when you did power up the system you were confronted with the Basic interpreter - which you had to use if you wanted to get anything done on the system. As a result, there was a high level of motivation to learn. Luckily for me my school actually taught programming -- Basic and Fortran on various machines (NCR and Apple]['s as well as CRT and teleprinter terminals to the county's PDP minicomputer)
Students need to be guided toward the resources - beyond all the hype surrounding the next high level language or development framework. To be effective they need to start at the beginning - crawl, walk then run. This is not occurring in our school system for programming with rare exception today.
Your arguments surrounding Python is just hand waving, smoke and mirrors. At the beginner level an interpreted language is far superior to a compiled language for providing immediate feedback. Furthermore, I disagree with your approach because it doesn't give kids the tools they need to explore the computer at the system level; 'flash' tells them nothing about how the computer works - and as the name implies is just glitz without substance. Programming is about the computer -- I think we've lost site of that in our ever spiralling quest to virtualize on top of it. Kids are memorizing libraries without learning what lays beneath.
Maybe what is needed is a new learning language that incorporates those basic capabilities of the early microcomputers - albeit a virtualized one today. A cpu and graphical simulator could underly an interpreted language so students could learn how to manipulate registers and the screen's graphics to foster exploration and creativity - similar to work done by Donald Knuth in his seminal Art of Computer Programming - the creation of "MIX assembly language," which runs on the hypothetical MIX computer. (Currently, the MIX computer is being replaced by the MMIX computer, which is a RISC version.) Software such as GNU MDK exists to provide emulation of the MIX architecture. We could take a similar approach at an even simpler level for beginners.
My message was not to 'let them off'. My message was that by playing the game by the rules of the securities market, by definition they will have to do what it takes to keep the value of their stock up or suffer the consequences.
If they are getting involved in lobbying activities, they are doing the same thing as other companies to provide a more favorable environment - which in turn keeps their stock value up.
Is that right or wrong? That depends upon the means they use to get their ends - which is the crux of this discussion.
The simple fact that shares were issued creates a pressure on the people running the show to maintain or increase the value of those shares - in fact I would say there is more motivation when the people running the show own controlling stake in the shares. Not only do stocks raise money for the company upon their issuance, banks watch stock performance to determine the viability of the endeavor, thus allowing the company to have access to new sources of money - needed in order to grow - or to increase thier margin with the stock exchange.
If they don't maintain the value of the shares - regardless of the type of share - the company will be beholden to lay out a significant sum of money if confidence drops and everyone dumps their shares on the market(probably exaserbating an already bad situation).
Whether shareholders overtly put pressure on a company to increase value or not is irrelevant. The simple fact that a company issued stock that is circulating and publicly traded creates pressure on a company to increase value or risk margin calls that can impact the company's bottom line - or in extreme cases lead to insolvency.
Once Google joined the system, by issuing stock, they did become part of that larger economic struggle - and if they don't maintain the stock value they will be beholden to the stock exchange for significant sums if their stock tanks. If they don't maintain the value of their stock - then they shouldn't have joined the game to begin with because it will cost them if they fail. Ipso facto they will attempt to maintain and increase their stock value.
Python comes installed on most Linuxes, and there is a binary installer for windows - I've installed it on various machines and linux and windows was the easiest by far. So I think your argument that it needs additional package setup is not correct - unless you are loading it on some machine that no average consumer uses.
From a beginner's perspective you don't need to do anything but load it and start using it:
$ python Python 2.3.5 (#1, Nov 30 2005, 10:43:26)... >>> a="Hello World" >>> print a Hello World >>> for x in range(1,10,1):... print a... Hello World Hello World Hello World Hello World Hello World Hello World Hello World Hello World Hello World >>> b=0 >>> while b >> if b == 10:... print a... else:... print "b="+str(b)... Hello World >>> print b 10
etc.
Python is perfectly capable of teaching basic concepts at a beginning level - and makes a powerful language when they are ready to dig deeper.
He missed ZOPE, with such add-ons as the Plone content management system, becomes very competitive in the RIA space. Zope uses the object database ZODB. Zope is written mostly in Python - and uses python as its development language of choice (although you can use others).
Been using it for years -- it is stable as a rock, and Version 3 is looking very cool. If you love Python, then Zope is the development framework to use imho.
I've bought at least two copies of grace jones' slave to teh rhythm, yet don't at present have one because the cassette and the cd both wore out.
A) If you take care of a CD - it will last as long as you can effectively use it. I have CDs from the '80s that are still playing strong.
B) Why didn't you burn your CD to MP3 before it became too damaged to play? Once you have it in an electronic medium, you can back it up off line, and access it through different devices. Then you can keep your CDs in a clean dry environment where they can be your ultimate backup - for as long as you live.
C) By buying 2 copies of the album - you just gave the music industry middle men twice what they deserve (and maybe more than that). The artist usually doesn't take home a large percentage of record sales.
I would have the added stipulation that the software on any hardware system be standards based...
Okay - who's standards?
You have to bare in mind interoperability. It is all well and good you want to standardize on 'Docbook' format for your documents - but might be a problem if you attempt to share these documents with your Microsoft Office-centric partners who have no desire to change.
I know a business that replaced several super computers with clusters.
1. They saved money (they were paying $100,000 per month for support - once the cluster was in place they managed their own support at a fraction of that cost).
2. They saved time (jobs that took the super computer 48 hours to complete, were completed by the new cluster of Linux machines interconnected with fiber in 1 hour). So they could do more work.
From my own experience, using mainframe equipment has drawbacks - the major one being a shared backplane that can become a single point of failure for the whole system. On the other hand, if a workstation/server motherboard goes bad, you can take it out of the cluster, replace or repair it without making a significant impact on operations.
Orion is the name for the ship type. The name of each ship will be different - so you could have an Orion 'class' interplanetary space ship named 'Enterprise' etc...
Sounds like launchd is a wrapper for the core unix init/cron services - that you could access directly if you wanted to. Best of both worlds is not a bad thing...
I've always wanted an IPod...but my wife has been talking about satellite radio - so I took a look at what was available.
Not only do they provide a satellite reciever, but the units also come with USB connectivity and an MP3 player capability - in additiona to docking stations for automobiles. You can upload your music to the machine and listen to it, and you can save satellite programming as MP3 for later download (reverse pod-cast?).
She is particularly interested in satellite radio because she doesn't care to sit down and find pod-casts, or music online --- she just wants to spin a dial and get a selection of music and talk radio on the fly. She also saw the benefit of being able to save what was currently playing to share with me - so she doesn't have to remember all the details to harange me by word-of-mouth later on - she can just put it on my multimedia server on my upload directory - so my player can automagically slirp it up (oh joy...).
At least I know what to get her for her birthday this year. I think I'll stick with an MP3 player myself - when I save enough money to get one.
Slashdot Mirror
Kawahee, you must not be a programmer. If you were, you would know that one of the key aspects of an effective programmer is laziness. Developers manifest this quality by spending inordinant hours researching (or less successfully asking /.), downloading and testing various tools built by other programmers in order to find the 'right thing'. If, and only if they can not find a tool that does a very high percentage of what they want, then they will break down and spend the time to build it themselves. More commonly, they will use the 'high percentage' tool, and build a smaller add-on to capture that last few percentage points.
This, of course, assumes the developer does not work in a shop where the choice of tools is dictated by management.
Under X windows (linux/unix - kde, gnome, motif etc...) -
1. Highlight text with left mouse button.
2. Go to new location and click middle mouse button to paste.
Voila! Your text is copied and pasted - probably the most efficient method available.
The DOS/Windows command line comes close:
1. Highlight text with left mouse button.
2. Right click in window to copy.
3. Right click in window to paste.
But this only works inside the CLI. If you want to take it out to another window, you are still stuck using a windows shortcut (CTRL V) or the right click menu 'paste' option.
With new complex applications running on platform architectures that are constantly changing (for such things as convergent applications), decentralizing your platform would be insane.
Your costs to change the system would go up. Your security would be exponentially more costly, and your telemetry and other time sensitive aspects of the management solution would go through the roof!
I would also hazzard a guess that removal of the datacenter would probably violate SOX (it would probably be a bad thing for Johnny in accounting to have physical access to the server housing his data).
Because typing is geek chic.
Removing your fingers from the home row on the keyboard takes too much time; quicker to memorize keyboard combinations to do what you need.
Whoa partner! You're using an awful broad brush to paint your picture.
Debian, Slackware, Ubuntu and other distros are rock solid and continually maintained.
Emacs, Python, Firefox, Open Office are some very big applications that are well maintained.
So - while some OSS projects may have serious maintenance problems, you can't say that for all of them. Additionally, I wouldn't be so quick, as some have, to believe proprietary software is any better maintained; I would argue just the opposite.
Absolutely -
The direction MS is going with windows will definitely not fill that niche - particularly for businesses that have to be cost concious and can't blow thousands or millions of dollars on new installations that support the latest and greatest from MS.
I wouldn't load Win98 on a 486 66 - even in 1998 - and as I mentioned that system only ran DOS before I got it. I do believe you can build a minimal Linux system on a Pentium 100 or better system that can appear just as full featured on the GUI as Win98 systems. The problem is no one in the Linux arena has that as a goal, that I am aware of (educate me if there are - as an aside, I do know there is a windows work-alike project called ReactOS that is going for binary equivalence to run win32 apps - and already has many native windows apps - MS Powerpoint, Adobe Photoshop etc.. - that will run on it).
To start such a project in Linux, you could list all the things you want to see on the system from an interface and tools perspective that a typical Win98 user would expect - and build to that standard. You would reach into your embedded systems bag of tricks to build highly tuned kernels that are fast, efficient, with a small footprint designed for those low-end processors. You would tweak the window manager to be more 'windows-like' (start with fvwm'95 or other lightweight window managers). On top of that you would add all the requirements for strange drivers - again the embedded approach would have merit - minimalistic and targeted. Finally, you would probably need to load WINE or Crossover Office in order to run ancient DOS and Windows apps that interface with the hardware or data produced by it - unless, of course, you could reverse engineer these applications and build them natively as a linux binary. I can see why this appears at first glance to be a daunting task.
Unfortunately, I haven't seen anything on the radar that targets this - which could be a missed opportunity for integrators - and a danger for those Win98 systems still attached to the network - as security erodes, or even those doing critical tasks off the net could be lost to corruption as drives fail over time - OEM installation disks notwithstanding. At some point these businesses will have to make a move - either replacing all the old gear, or spending the time building an open source solution that can have a lifespan longer than dictated by a vendor.
Not an easy problem; thankfully I don't have that problem to deal with - hence why I thought I could spend a few moments pointing out some alternatives.
I hacked the zipslack Slackware distro to load on my Acer 486 66mhz laptop (20 MB ram, 500 MB hard drive); I have a parallel port iOmega 100 MB Zipdrive I used to hold the needed packages for a minimal system, and used the regular boot disk with the Parallel port root disk - to gain access to the drive.
Once I got the minimal system running, I transfered the packages I required from the distro via the zipdrive, and used installpkg to flesh out the applications I wanted from the regular distro.
As a finishing touch I built the TWIN text based based window manager - which is implimented in curses and emulates X11 interfaces - on my workstation, then transfered the working window manager over. X.org would not fit on the system given the limitations of space and video capability of that system. It runs great - much better than the DOS 4 that was single-tasking on it when I got it.
So it is possible to breath new life into very marginal (by today's standards) machines.
The big problem today is most highschools do not provide computer programming as part of their cirriculum. Instead they teach students to run MS Windows, 'Word', 'Excel', and browsers, maybe a graphics art class using photoshop, and if they are real lucky a computer hardware class where they learn to build a PC from parts, troubleshoot components and maybe build an electronic circuit from parts - none of which helps students become programmers. So they are on their own, for the most part, and enter CS degree program without a clue - if they enter it at all (CS major enrollments are down for a number of factors including offshoring - but largely due to lack of opportunities in CS in public schools).
Back in the 'good old days' - if you were lucky enough to have a microcomputer (the TI 99a my parents bought for me in 1981 was a whopping $2000!! - the better part of the value of a car back then) or access to microcomputers in a fledgling 'computer science' class (I was lucky enough to also attend the first CS class provided at my highschool) - when you did power up the system you were confronted with the Basic interpreter - which you had to use if you wanted to get anything done on the system. As a result, there was a high level of motivation to learn. Luckily for me my school actually taught programming -- Basic and Fortran on various machines (NCR and Apple]['s as well as CRT and teleprinter terminals to the county's PDP minicomputer)
Students need to be guided toward the resources - beyond all the hype surrounding the next high level language or development framework. To be effective they need to start at the beginning - crawl, walk then run. This is not occurring in our school system for programming with rare exception today.
Your arguments surrounding Python is just hand waving, smoke and mirrors. At the beginner level an interpreted language is far superior to a compiled language for providing immediate feedback. Furthermore, I disagree with your approach because it doesn't give kids the tools they need to explore the computer at the system level; 'flash' tells them nothing about how the computer works - and as the name implies is just glitz without substance. Programming is about the computer -- I think we've lost site of that in our ever spiralling quest to virtualize on top of it. Kids are memorizing libraries without learning what lays beneath.
Maybe what is needed is a new learning language that incorporates those basic capabilities of the early microcomputers - albeit a virtualized one today. A cpu and graphical simulator could underly an interpreted language so students could learn how to manipulate registers and the screen's graphics to foster exploration and creativity - similar to work done by Donald Knuth in his seminal Art of Computer Programming - the creation of "MIX assembly language," which runs on the hypothetical MIX computer. (Currently, the MIX computer is being replaced by the MMIX computer, which is a RISC version.) Software such as GNU MDK exists to provide emulation of the MIX architecture. We could take a similar approach at an even simpler level for beginners.
My message was not to 'let them off'. My message was that by playing the game by the rules of the securities market, by definition they will have to do what it takes to keep the value of their stock up or suffer the consequences.
If they are getting involved in lobbying activities, they are doing the same thing as other companies to provide a more favorable environment - which in turn keeps their stock value up.
Is that right or wrong? That depends upon the means they use to get their ends - which is the crux of this discussion.
My argument stands.
The simple fact that shares were issued creates a pressure on the people running the show to maintain or increase the value of those shares - in fact I would say there is more motivation when the people running the show own controlling stake in the shares. Not only do stocks raise money for the company upon their issuance, banks watch stock performance to determine the viability of the endeavor, thus allowing the company to have access to new sources of money - needed in order to grow - or to increase thier margin with the stock exchange.
If they don't maintain the value of the shares - regardless of the type of share - the company will be beholden to lay out a significant sum of money if confidence drops and everyone dumps their shares on the market(probably exaserbating an already bad situation).
Whether shareholders overtly put pressure on a company to increase value or not is irrelevant. The simple fact that a company issued stock that is circulating and publicly traded creates pressure on a company to increase value or risk margin calls that can impact the company's bottom line - or in extreme cases lead to insolvency.
Once Google joined the system, by issuing stock, they did become part of that larger economic struggle - and if they don't maintain the stock value they will be beholden to the stock exchange for significant sums if their stock tanks. If they don't maintain the value of their stock - then they shouldn't have joined the game to begin with because it will cost them if they fail. Ipso facto they will attempt to maintain and increase their stock value.
Python comes installed on most Linuxes, and there is a binary installer for windows - I've installed it on various machines and linux and windows was the easiest by far. So I think your argument that it needs additional package setup is not correct - unless you are loading it on some machine that no average consumer uses.
... ... print a ... ... print a ... else: ... print "b="+str(b) ...
From a beginner's perspective you don't need to do anything but load it and start using it:
$ python
Python 2.3.5 (#1, Nov 30 2005, 10:43:26)
>>> a="Hello World"
>>> print a
Hello World
>>> for x in range(1,10,1):
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
>>> b=0
>>> while b >> if b == 10:
Hello World
>>> print b
10
etc.
Python is perfectly capable of teaching basic concepts at a beginning level - and makes a powerful language when they are ready to dig deeper.
This assumes that your employer gives you the option to agree or not.
You are making an unnecessary distinction. They are all slimeballs - some more agreeable to your own sensibilities than others.
As for Google - the moment they became a publicly held company, they became pawns to the larger economic struggle to maintain and improve shareholder value.
Deal.
;)
As a matter of fact, there are a bunch of GUI libraries being or already developed for python.
The point is, you don't have to drink the purple koolaid if you don't want to.
He missed ZOPE, with such add-ons as the Plone content management system, becomes very competitive in the RIA space. Zope uses the object database ZODB. Zope is written mostly in Python - and uses python as its development language of choice (although you can use others).
Been using it for years -- it is stable as a rock, and Version 3 is looking very cool. If you love Python, then Zope is the development framework to use imho.
How do you know 'most of us' are human beings? ;)
Until MS breaks some interface, as they have done in the past.
Python and Tk give you everything you really need to develop GUI apps on any system. Throw in Zope and you can web-enable it.
KISS - debugging complex apps is difficult enough without adding another layer of cruft on top of if - just so I can use some 'gee-whiz'
A) If you take care of a CD - it will last as long as you can effectively use it. I have CDs from the '80s that are still playing strong.
B) Why didn't you burn your CD to MP3 before it became too damaged to play? Once you have it in an electronic medium, you can back it up off line, and access it through different devices. Then you can keep your CDs in a clean dry environment where they can be your ultimate backup - for as long as you live.
C) By buying 2 copies of the album - you just gave the music industry middle men twice what they deserve (and maybe more than that). The artist usually doesn't take home a large percentage of record sales.
Okay - who's standards?
You have to bare in mind interoperability. It is all well and good you want to standardize on 'Docbook' format for your documents - but might be a problem if you attempt to share these documents with your Microsoft Office-centric partners who have no desire to change.
That is an oxymoron.
I know a business that replaced several super computers with clusters.
1. They saved money (they were paying $100,000 per month for support - once the cluster was in place they managed their own support at a fraction of that cost).
2. They saved time (jobs that took the super computer 48 hours to complete, were completed by the new cluster of Linux machines interconnected with fiber in 1 hour). So they could do more work.
From my own experience, using mainframe equipment has drawbacks - the major one being a shared backplane that can become a single point of failure for the whole system. On the other hand, if a workstation/server motherboard goes bad, you can take it out of the cluster, replace or repair it without making a significant impact on operations.
Decentralization is good imho.
Quantity != Quality
Orion is the name for the ship type. The name of each ship will be different - so you could have an Orion 'class' interplanetary space ship named 'Enterprise' etc...
Sounds like launchd is a wrapper for the core unix init/cron services - that you could access directly if you wanted to. Best of both worlds is not a bad thing...
I've always wanted an IPod...but my wife has been talking about satellite radio - so I took a look at what was available.
Not only do they provide a satellite reciever, but the units also come with USB connectivity and an MP3 player capability - in additiona to docking stations for automobiles. You can upload your music to the machine and listen to it, and you can save satellite programming as MP3 for later download (reverse pod-cast?).
She is particularly interested in satellite radio because she doesn't care to sit down and find pod-casts, or music online --- she just wants to spin a dial and get a selection of music and talk radio on the fly. She also saw the benefit of being able to save what was currently playing to share with me - so she doesn't have to remember all the details to harange me by word-of-mouth later on - she can just put it on my multimedia server on my upload directory - so my player can automagically slirp it up (oh joy...).
At least I know what to get her for her birthday this year. I think I'll stick with an MP3 player myself - when I save enough money to get one.