Copernic desktop search let's you select the extension of the files to search for each category. It has a very nice and polished interface and personally like it more than google or msn or yahoo desktop searches.
no problem:)
And yes, Arabic, Persian and languages like that read from right to left. However, the funny part is that the numbers are read from left to right! Now, the bug with FF is that when you write 314 (with arabic characters, that is) it renders it correctly as 314 but when you write 3.14 it swaps the two parts and render it 14.3 which can be a bit annoying if you have lots of such numbers in a text.
Oh, shit! Rendering of non-English pages is much more important than preventing my workstation from being ass raped by a malicious page that masquerades as a useful search result...
Did I say that? I was just answering this phrase: "...Firefox will render any site closer to what the site's html/xml code is specifying..."
I just brought one simple example that FireFox sometimes do not render something correctly. You have a problem with that?
I typed my example using latin numbers but the bug is with the arabic numbers (that I could not use corrctly with./ post editor!). And you are right, the encoding of the page should not change the order of characters. Firfox simply shows such number in arabic "always" with reversed characters round the decimal point!
And before you get angry again, please note that I'm a FireFox user myself, and I know this bug is there and yet I'm continuing to use for OTHER benefits it has.
Well, generally I agree with you. However, when it comes to correctly rendering UTF-8 pages, specially with Arabic characters, firefox has some very well known bugs that have not been fixed now for ages. The most annyoing one is a bug in rendering arabic decimal number: It shows all numbers like 1.4 as 4.1! Of course, IE renders such pages perfectly.
Is it me, or can they just take the Safari_Khtml and work from that? Seems easier to me...
hmmmm, guess it's you;)
Safari is based on KHTML 3.1 so it is based on some older branch. Taking it as the starting point means abondoning KHTML 3.2 in favour of a code base that some people thing is not up to the quality standards of the rest of the project.
The entire 127/8 is reserved for loopback purposes
Actually, 128 is not reserved for that purpose. Consider this:
C:\>ping www.epfl.ch
Pinging www3.epfl.ch [128.178.50.137] with 32 bytes of data:
Reply from 128.178.50.137: bytes=32 time=XXms TTL=XXX
Lots of code that copy a pointer to a long?? As a C++ programmer, I don't see any sane reason in copying a pointer to a long or any other scaler types. If there are codes around using bad practices like that, I wouldn't care if they'd get broken.
I understand the reasons why people would like to have a 64 bits long and a 32 bits int by default, but copying a poitner to a long is not one of the valid ones.
Well, yes, it is possible. There are other (real) possibilities too. However, the question is as always how much you want to pay for how much security. I didn't claim this is a method to answer all possibilities. Maybe it is possible to find such a solution but surely it won't be cheap.
This method has been used on a single computer, without the possibility to access to a remote backup server. Also, size of the data is relatively small (only a few gigabytes) so an expensive fault-torlent system like a RAID5 server with a dual power supply would be an overkill. This method is usefull for users at home or small offices where even cost of a second hard disk is an issue to discuss and think of:)
I was just tring to say that a Disk-to-Disk backup system can be a little bit more complex than just a simple copy command. That's all.
What's wrong with VSS? This is a single machine, and I use Visual Studio to write C++ code and MATLAB for developing DSP algorithms. Both IDEs integrate very well with VSS and the only moment I even feel the presence of VSS is when I create a project and add check a check box to add it to the source control.
I can also use CVS with the same level of integration in these tools? I don't know. Maybe yes, but why should I care if it is already working with no problem.
Sorry, but I'm not part of any cult. I don't care who has written this too. If I have it and it works and can solve my problem, then I use it.
Sure, you're right. In my a case a true RAID would be overkill. This solution of mine is really just a poor man's RAID!
And yes, I use a version control system also. The database of the version control system is part of what I mirror in the daily basis.
I said RAID has no advantage is this regard. I know where to use and where not to use the RAID. You'd better learn when to use what tool for what problem. This is a system used on single workstation, not a company's webserver or whatever!
With your scenario, such disaster would at most set me back for one day worth of work. Fine, given the probabilities of it, I'm happy that my very inexpensive solution (i.e very low overhead) is quite acceptable. And guess what, on you beloved RAID, if I by mistake ruin or delete a file (which happens on a development workstation quite a lot), I'll lose it for ever, but with my method, I have at least 12 hours to recover it!
In that situation RAID is better than once-a-day backups onto the same devices.
RAID has no advantage in this regard. As I said, I make two backup copies: one on each drive. If there is a hardware failure, there is always a safe copy on the other drive.
I use this method exactly to avoid the hardware problems without the high overhead of a real RAID system. About versioning, well, I have all my codes in VSS and the whole VSS database is also part of what I mirror in backup folders.
Yes, I could as well configure the hard disks in a RAID configuration. The advantage would be instant mirroring of everything written on the disks. However, I don't really need that. I would lose lots of disk space in RAID. I have 2x160GB hard disks. In a RAID configuration, I woud have only 160GB free space (ie. 160G / 160G = 100% overhead). In my case, the total size of the "important data" is now only round 8GB. I have two copies of the backup, so I'm wasting only 2x8GB = 16GB for backup. The backup directories are actually NTFS compressed folders so the real space uesd for each one is just 5GB instead of 8GB or 10G in total. So, now I have only 10G / 320G = 3.1% of overhead!:)
Disk-to-disk backup? In fact, I (ie. my computer!) do it every night. Simple copy command? I think that does not cut it. I'm in a tight development cycle and each day write a lot of code, documents and receieve/generate lots of data files. I need to back up all important data but surely I don't need to make backup of the executable files, temp files, OS system files and such.
The solution that I use is simple: I have two hard disks in my computer. The files that I need to have back-up from, are scattered on these two drives. Now, I have made a BackUP directory on each one of these drives and put a copy of all important files in them. So, I have 3 copies of every important files: the original, and two back-ups. In case a hard-disk goes banana, I always have a copy of all important files on the other one.
I run the back-up every night. Just need to copy the files that has been changed or the whole new directories made during the day.
So the problem is: I need two desinations for each source. I need to be able to select which directories or even which files to back-up (or not to backup) and I need to check which files have been changed or which new files (or directories) have been created. I need to be able to schedule the back ups for midnight and I need to forget about all these details in practice as I have to focus on my work:)
How I did it? Well, I tried a script in the beginning but found it difficult to manage over the time and it was very tedius. Now I use SyncBack which is a freeware program with all these features that I need (and more! like FTP and compression to Zip, etc.). QED.
I also believe the fundamental problem is that managers these days (in many cases) no longer come from the ranks and are not engineers. So, they do not always understand what is involved in 1) building the codebase 2) testing code base 3) proper interface design 4) end user testing 5) documentation 6) making sure it does not suck.
Although you're point is valid, but could it be that 95% of IT projects have some changes in the specs AFTER the schedule has been made? Is it just me or do you also see a correlation here?
Yes, he's just resting but yet he has the top rating in the chess and many consider him as the greatest player in the world. He is in a good shape and although he has lost his last game in the Linares tournament, but has won the tournament.
Chess like any other game/sport needs sponsors, fans and supporters and Kasparov has been a great name attracting many. His mere presense in a tournament would mean big support and big moneys for Chess. Other active Grandmasters, though not very far from him in theory, could not yet attract so much publicity, support and money to Chess. So I think to the Chess world, his retirement is a very sad news.
Well, although all your suggested methods may work but personally I think the best solution for my PhD example is to make a website and sell only the binary version of his plug-in. The level of support can be the absolute minimal: make a BB and let the users support each other! You just go there regularly to answer some questions and get some feedback. This works and I didn't make this example in the air: there are 2 or 3 ISO noise removal products on the net (use google if you like) that are the best in the class, closed-source and follow the same one man army pattern I described. And guess what, they are selling very well!
You want an answer to how EVERYONE can make money with opensource?
Not everyone but yes, lots of people. My take is the solutions offered here and in other places (selling support or specialized services) would only work for a small number of companies working on small sets of programs. But I think the real question is how open-source methology can be adopted by a large number of developers.
So in short, trolltech makes money on its closed-source products, not the open-sourced ones, right? And tell me, who pays the money anyway? Some other company that will release the source code of this spcialized, paid so much for version of the program or the one that keeps the source closed? It seems that money comes anyway from a closed-source program.
Seems that you didn't get my point. And also it seems that you think software means only mailservers, webservers and such. I know RedHat (just an example) also write LOTS of things, but comparing to what they distribute, maybe it is only round 1% of the whole thing. This model works, because thosusands of people contribute freely to different open source projects for FREE and only a few companies sell the support. Nobody pays those thousands of people anything. You say, yeah, they did it for free so they don't expect to receive money and I say that's my point! What if I'm a developer and I want to make money and I'm not a big company like RedHat? And what if my PhD is in signal processing and I can write the results of years of my studies and research in form of super-cool image noise reduction algorithms (and have no clue how to write a mailserver). Is your suggestion for me is to make and open-source program, write my algorithm in the form of a PhotoShop plugin and earn money by selling support? Do you really think that gonna work? And what if I just sell the binaries? Which one will make more sense for a developer or an small company?
You miss one big point: not all programs need lots of support. For example, tell me how much RedHat asks for the support of his HUGE boundle of programs they have on their CDs? Let's say 70$ per year. Now, you, the developer, have just written ONE single program. How low can you make the price of the support for your single program? 1$ per year? Would it be enough for you? How many people would buy this support? And why I, the customer, would ever buy support from you when I can pay 70$ to RedHat and buy support for 100s of programs?
It seems the model you have described is only going to help big companies make more money and put the real, small, developers out of business.
Good point. It seems that those companies basically are selling support for the programs they have not written themself.
Frankly, I don't think this is going to be the answer to the question: How can you make money by writing open-source programs.
Guess you've never seen Visual Assist. No, you can't do MOST of what it does by simply using VIM. First, Visual Assist is not a text editor, it is a plugin for an already very good programmer's editor: Visual Studio. But it goes far beyound normal syntax highlighting. Why? Beacause Visual Assist is aware of the syntax of C/C++/VB/C#... and acts accordingly. For example, it can diffrentiate between local can globalr variables. The functions you have in your code and the functions you have used from other libraries. When you use it with existing libraries with source (like Boost for example), it will parse the whole thing and now Suggestion lists and spell checker will be aware of the new functions, methods and classes. So you can't simulate it with just a hand made syntax file, as suggestion lists and spell checker are project dependant. It also learns from the patterns you use more often in your code and sorts the suggestions in a way that the probability that you'd need to scroll down the list of the suggestions becomes lower and lower. Spell-checker nows the name of the local variables, included functions and namespaces so even before running the compiler, you know that you have not miss-spelled that function call. All in all, Visutal Assist is in a whole different class than VIM.
The point about this program is not about using DX 9.0c accelerator cards to draw the GUI, but to accelerate the effects engine of the program using pixel shader 3.0 insturctions. A modern graphics card can run filters and nonlinear effects MUCH faster than the faster CPU. Now, with nVidia cards you have 32 bits floating point numbers for each of the RGBA channels of a pixel which is quite enough for many signal processing algorithms while ATI's 24 bits floting point numbers are a bit limiting.
Well, some countries are very sensitive about such issues. Even people can be very sensitive about it. Take this recent example: You know that some Arab countries insist on using the name "Arabian Gulf" to call whan we know as "Persian Gulf". Recently, after mentioning this second name in some national geography publications, a large group of Persian weblogs and sites helped making a google bomb. Try searching for "Arabian Gulf" in google and select the first results and see it for yourself.
I agree. The design goals of this device is clearly are: Providing minimum requirments to access and browse the net, and do basic word processing and that's all! It should not require any service, patches for a long time. Many slashdotters many nor realize that design of such a system is quite different from normal PCs and is a completely different beast. You need a system to do something very basic, only do this job and do it right with no future need for any chagnes. This is logical to lock it in a way that making any changes to it be very difficult. Not only to protect it from the users, but also from all unknown possible future threats on the net. You can imagine millions of such simple machines round the world and far from any service center can be a very good target for spammers, black-hats and anyone who needs a hord of zobie machines.
How would you design such a thing if you needed to do it with such a tight price limit?
Slashdot Mirror
Copernic desktop search let's you select the extension of the files to search for each category. It has a very nice and polished interface and personally like it more than google or msn or yahoo desktop searches.
no problem :)
And yes, Arabic, Persian and languages like that read from right to left. However, the funny part is that the numbers are read from left to right! Now, the bug with FF is that when you write 314 (with arabic characters, that is) it renders it correctly as 314 but when you write 3.14 it swaps the two parts and render it 14.3 which can be a bit annoying if you have lots of such numbers in a text.
I just brought one simple example that FireFox sometimes do not render something correctly. You have a problem with that?
I typed my example using latin numbers but the bug is with the arabic numbers (that I could not use corrctly with
And before you get angry again, please note that I'm a FireFox user myself, and I know this bug is there and yet I'm continuing to use for OTHER benefits it has.
Well, generally I agree with you. However, when it comes to correctly rendering UTF-8 pages, specially with Arabic characters, firefox has some very well known bugs that have not been fixed now for ages. The most annyoing one is a bug in rendering arabic decimal number: It shows all numbers like 1.4 as 4.1! Of course, IE renders such pages perfectly.
Safari is based on KHTML 3.1 so it is based on some older branch. Taking it as the starting point means abondoning KHTML 3.2 in favour of a code base that some people thing is not up to the quality standards of the rest of the project.
C:\>ping www.epfl.ch
Pinging www3.epfl.ch [128.178.50.137] with 32 bytes of data:
Reply from 128.178.50.137: bytes=32 time=XXms TTL=XXX
Lots of code that copy a pointer to a long?? As a C++ programmer, I don't see any sane reason in copying a pointer to a long or any other scaler types. If there are codes around using bad practices like that, I wouldn't care if they'd get broken.
I understand the reasons why people would like to have a 64 bits long and a 32 bits int by default, but copying a poitner to a long is not one of the valid ones.
Well, yes, it is possible. There are other (real) possibilities too. However, the question is as always how much you want to pay for how much security. I didn't claim this is a method to answer all possibilities. Maybe it is possible to find such a solution but surely it won't be cheap. :)
This method has been used on a single computer, without the possibility to access to a remote backup server. Also, size of the data is relatively small (only a few gigabytes) so an expensive fault-torlent system like a RAID5 server with a dual power supply would be an overkill. This method is usefull for users at home or small offices where even cost of a second hard disk is an issue to discuss and think of
I was just tring to say that a Disk-to-Disk backup system can be a little bit more complex than just a simple copy command. That's all.
What's wrong with VSS? This is a single machine, and I use Visual Studio to write C++ code and MATLAB for developing DSP algorithms. Both IDEs integrate very well with VSS and the only moment I even feel the presence of VSS is when I create a project and add check a check box to add it to the source control.
I can also use CVS with the same level of integration in these tools? I don't know. Maybe yes, but why should I care if it is already working with no problem.
Sorry, but I'm not part of any cult. I don't care who has written this too. If I have it and it works and can solve my problem, then I use it.
Sure, you're right. In my a case a true RAID would be overkill. This solution of mine is really just a poor man's RAID!
And yes, I use a version control system also. The database of the version control system is part of what I mirror in the daily basis.
I said RAID has no advantage is this regard. I know where to use and where not to use the RAID. You'd better learn when to use what tool for what problem. This is a system used on single workstation, not a company's webserver or whatever!
With your scenario, such disaster would at most set me back for one day worth of work. Fine, given the probabilities of it, I'm happy that my very inexpensive solution (i.e very low overhead) is quite acceptable. And guess what, on you beloved RAID, if I by mistake ruin or delete a file (which happens on a development workstation quite a lot), I'll lose it for ever, but with my method, I have at least 12 hours to recover it!
I use this method exactly to avoid the hardware problems without the high overhead of a real RAID system. About versioning, well, I have all my codes in VSS and the whole VSS database is also part of what I mirror in backup folders.
Yes, I could as well configure the hard disks in a RAID configuration. The advantage would be instant mirroring of everything written on the disks. However, I don't really need that. I would lose lots of disk space in RAID. I have 2x160GB hard disks. In a RAID configuration, I woud have only 160GB free space (ie. 160G / 160G = 100% overhead). In my case, the total size of the "important data" is now only round 8GB. I have two copies of the backup, so I'm wasting only 2x8GB = 16GB for backup. The backup directories are actually NTFS compressed folders so the real space uesd for each one is just 5GB instead of 8GB or 10G in total. So, now I have only 10G / 320G = 3.1% of overhead! :)
Disk-to-disk backup? In fact, I (ie. my computer!) do it every night. Simple copy command? I think that does not cut it. I'm in a tight development cycle and each day write a lot of code, documents and receieve/generate lots of data files. I need to back up all important data but surely I don't need to make backup of the executable files, temp files, OS system files and such. The solution that I use is simple: I have two hard disks in my computer. The files that I need to have back-up from, are scattered on these two drives. Now, I have made a BackUP directory on each one of these drives and put a copy of all important files in them. So, I have 3 copies of every important files: the original, and two back-ups. In case a hard-disk goes banana, I always have a copy of all important files on the other one. I run the back-up every night. Just need to copy the files that has been changed or the whole new directories made during the day. So the problem is: I need two desinations for each source. I need to be able to select which directories or even which files to back-up (or not to backup) and I need to check which files have been changed or which new files (or directories) have been created. I need to be able to schedule the back ups for midnight and I need to forget about all these details in practice as I have to focus on my work :)
How I did it? Well, I tried a script in the beginning but found it difficult to manage over the time and it was very tedius. Now I use SyncBack which is a freeware program with all these features that I need (and more! like FTP and compression to Zip, etc.). QED.
Although you're point is valid, but could it be that 95% of IT projects have some changes in the specs AFTER the schedule has been made? Is it just me or do you also see a correlation here?
Yes, he's just resting but yet he has the top rating in the chess and many consider him as the greatest player in the world. He is in a good shape and although he has lost his last game in the Linares tournament, but has won the tournament.
Chess like any other game/sport needs sponsors, fans and supporters and Kasparov has been a great name attracting many. His mere presense in a tournament would mean big support and big moneys for Chess. Other active Grandmasters, though not very far from him in theory, could not yet attract so much publicity, support and money to Chess. So I think to the Chess world, his retirement is a very sad news.
Not everyone but yes, lots of people. My take is the solutions offered here and in other places (selling support or specialized services) would only work for a small number of companies working on small sets of programs. But I think the real question is how open-source methology can be adopted by a large number of developers.
So in short, trolltech makes money on its closed-source products, not the open-sourced ones, right? And tell me, who pays the money anyway? Some other company that will release the source code of this spcialized, paid so much for version of the program or the one that keeps the source closed? It seems that money comes anyway from a closed-source program.
Seems that you didn't get my point. And also it seems that you think software means only mailservers, webservers and such. I know RedHat (just an example) also write LOTS of things, but comparing to what they distribute, maybe it is only round 1% of the whole thing. This model works, because thosusands of people contribute freely to different open source projects for FREE and only a few companies sell the support. Nobody pays those thousands of people anything. You say, yeah, they did it for free so they don't expect to receive money and I say that's my point! What if I'm a developer and I want to make money and I'm not a big company like RedHat? And what if my PhD is in signal processing and I can write the results of years of my studies and research in form of super-cool image noise reduction algorithms (and have no clue how to write a mailserver). Is your suggestion for me is to make and open-source program, write my algorithm in the form of a PhotoShop plugin and earn money by selling support? Do you really think that gonna work? And what if I just sell the binaries? Which one will make more sense for a developer or an small company?
You miss one big point: not all programs need lots of support. For example, tell me how much RedHat asks for the support of his HUGE boundle of programs they have on their CDs? Let's say 70$ per year. Now, you, the developer, have just written ONE single program. How low can you make the price of the support for your single program? 1$ per year? Would it be enough for you? How many people would buy this support? And why I, the customer, would ever buy support from you when I can pay 70$ to RedHat and buy support for 100s of programs? It seems the model you have described is only going to help big companies make more money and put the real, small, developers out of business.
Guess you've never seen Visual Assist. No, you can't do MOST of what it does by simply using VIM. First, Visual Assist is not a text editor, it is a plugin for an already very good programmer's editor: Visual Studio. But it goes far beyound normal syntax highlighting. Why? Beacause Visual Assist is aware of the syntax of C/C++/VB/C#... and acts accordingly. For example, it can diffrentiate between local can globalr variables. The functions you have in your code and the functions you have used from other libraries. When you use it with existing libraries with source (like Boost for example), it will parse the whole thing and now Suggestion lists and spell checker will be aware of the new functions, methods and classes. So you can't simulate it with just a hand made syntax file, as suggestion lists and spell checker are project dependant. It also learns from the patterns you use more often in your code and sorts the suggestions in a way that the probability that you'd need to scroll down the list of the suggestions becomes lower and lower. Spell-checker nows the name of the local variables, included functions and namespaces so even before running the compiler, you know that you have not miss-spelled that function call. All in all, Visutal Assist is in a whole different class than VIM.
Well, some countries are very sensitive about such issues. Even people can be very sensitive about it. Take this recent example: You know that some Arab countries insist on using the name "Arabian Gulf" to call whan we know as "Persian Gulf". Recently, after mentioning this second name in some national geography publications, a large group of Persian weblogs and sites helped making a google bomb. Try searching for "Arabian Gulf" in google and select the first results and see it for yourself.
I agree. The design goals of this device is clearly are: Providing minimum requirments to access and browse the net, and do basic word processing and that's all! It should not require any service, patches for a long time. Many slashdotters many nor realize that design of such a system is quite different from normal PCs and is a completely different beast. You need a system to do something very basic, only do this job and do it right with no future need for any chagnes. This is logical to lock it in a way that making any changes to it be very difficult. Not only to protect it from the users, but also from all unknown possible future threats on the net. You can imagine millions of such simple machines round the world and far from any service center can be a very good target for spammers, black-hats and anyone who needs a hord of zobie machines. How would you design such a thing if you needed to do it with such a tight price limit?