All good suggestions. As I posted on a higher level, people focus too much on skins. Others I suggest:
* Make use of the Win Key! (Or some other replacement). It's really nice to have ONE keystroke that gets me almost everywhere I need to go under windows. Also, copy the Ctrl-Insert, Shift-Delete, combo's. * Merge better with XFree project. Wouldn't it be great if I could simultaneously configure KDE and X because they share similar options? * Somehow better integrate KDE with other parts of Linux so that similar devices don't have to be configured twice. On an earlier release, for instance, I had to define the backspace key both in X and under the console mode. * Better proxy support, ideally for all relevant applications--configured in one place. I really hate dealing with firewalls, and this might be a good place to do battle with those things. * Never make me edit a text file to configure a system setting. The less I have to know about what file to edit, where it is, and what the format is, the better and faster I'd be.
Can we all get off this theme/skin kick? Really, the whole point of the new release was better connectivity between applications. Better integration. People who worship command lines should focus more on how things work, not on how things look.
Since we can't expect every web person to consider good design, here's some future requirements for browsers:
* Provide an easy to reach on/off switch for irritants like large images animations and flash. * If I have flash turned off, don't tell me so every time (IE does this) * Allow me to turn off all pictures/animations/etc and click on the page to then selectively download just that one. * Allow a total popup ban, without some third party tool. * Don't allow the browser to become full screen, or let me turn this off. * Always give me the minimize, restore and close buttons, never let some web bastard take them away from me. * Let me specify the Java version I use. * Give me multiple proxy options--our's a frequently down for whatever reason, and I don't want to dig deep into options to manually switch them. And for that matter, help me figure out what proxy is being used, since I can't find that out easily. * Minimize the unknown information disclosed to the web server.
Maybe so, but they still should include some form of assembler. It doesn't much matter what platform, so long as they have a clue on what it really takes for the computer to do something. Too often, programmers write sloppy, inefficient, badly patterned code with their only justification being that CPU's keep getting faster, and memeory cheaper.
I agree that I'd be really nice to cut down on the footprint size, and power usage. But I don't need the cute ruby, saphire, boxes with the rounded shapes. For me, I'd rather have this:
A larger footprint that I would last several generations of components. Oh, and easy component accessability, which probably rules out the Dells, IBMs and Compaqs.
A Separate box for removable media that's at my finger tips, not on the floor. IBM had a model like this, (Stealth?) but alas, no one else liked it.
Built in surge protection.
Wheels.
Commonly used screws.
Reasonably priced. Not $10,000 for a bare rack. *Note the exageration*
I guess I'm unique in this. I just want something boring, unfashionable, ugly, but useful and upgradeable.
MSDN used to be on CD's (possibly still true today, I'm not sure) which would have been far faster then any website. But then everything has to be web based right? But maybe it's just bad site design. I too was put on hold for close to an hour for tech support on MS Access, and I eventually gave up, called the premium line anyway, and actually got someone. He attempted to solve my problem, but mostly told me I couldn't do what I wanted because it didn't fit some paradigm of theirs.
A better example:
If you want good technical support, and you're willing to pay, go with IBM. Unlike MS, you can actually look up those cryptic error messages and not get 18 billion hits irrelevant hits. Things you need are organized into standard "books"-- references, guides, intro's, etc. When you need to quickly look something up, you know where to go, for specific answers not reading an entire RFC or gobligook, or ad laden website. I'm sure there are exceptions, though.
Also, support is something of a separate profession. People are actually measured for timeliness, quality, number of defects closed, etc. Also, fixes to fixes are heavily looked down upon.
You get this, because you pay for it, (quite a bit I'm told). But you don't have to wait for a plethora of people to have the problem to get it fixed. You also don't have to learn the internals to fix it like you would in an open source model.
One bad thing on Open Source support:
One time I wanted to turn off that annoying beep that the commandline mysql client has. When I posted my question to a mysql group, not IRC, No one could tell me how. One person even had the guts to tell me it's there for a reason, and I shouldn't try to circumvent it, but just put up with it! In my case, I just want to use Mysql, I don't want to get the source, find out where the beep comes from, and fix it myself. Sadly, I think that's the response one should expect to get when people don't get paid to help. I'm sure other's have had better results, and I'm happy they have, but I haven't seen the good results to be predictable like they would be with a support contract (not from Microsoft). If you had the money to pay for a support contract for a piece of OSS, then you'd have a different story.
In most OS's, you wouldn't want to do away with the CLI unless the GUI was absolutely perfect to your liking, and you never needed to do hard maintenance with it. I don't know what OS qualifies for this. I personally love to use vim, and find that to be one of my favorite *nix applications (of course I also use the NT version). Also, automatic administrative tasks would be much harder if we didn't have a good CLI.
Conversely, I also use Visual SlickEdit, which uses a lot of point and click, and makes navigating massive programs much easier. Sometimes I use the autocomplete feature. It's also nice to see method signatures popup in a tooltip. Or expand and collapse classes with a click. Also, before you dismiss the value of a GUI, think of the number of times you wanted to do file operations on files within a directory or on a set of files from a search result. It's a royal pain to have to scroll back up to see the directory contents, or worse, have to reissue the directory command.
I should point out, that although I do use GUI based tools, I don't use the mouse very often. I suspect that the most competent Windows users use a lot of the keyboard shortcuts. The GUI, in this situation, gives us a menu of possibilities so we don't have to memorize commands, but we can still use the keyboard and be fast about it.
Finally, CLI don't have to be difficult. The designers could use more verbose and intuitive commands and use a consistent, non symbol heavy syntax. My favorite example is the old VM/CMS command line. Getting a list of files: FILE. Checking your mail: READERLIST. And they could be shortened via synonyms or the first x characters of the command uniquely identifying it. Commands commonly used keyword parameters so at least you didn't need to know the order. Plus, some programs used a menu based interface (though it was text).
My conclusion: I already have a set of interfaces I'm used to using. Please, don't make other's switch because eventually someone is going to force that on me, too. (Ie, making Quicken more like a web browser instead of using the old MDI interface). Also, whatever it was that the researcher wanted to replace the desktop metaphor with sounded confusing when I read it, so maybe it wouldn't be as intuitive to use it.
I must be missing something. If you eliminate ACK packets, and reduce the amount of regular packets to "symbol" based packets, how can that possibly generate more traffic? This is an honest question, I'm not trying to insult anyone.
What would you define an OS to include. If it includes utilities such as a defragmenter that is also sold separately by another vendor, is that really so bad? In essence, how much of Windows, the OS itself and packaging would you remove to make it not violate antitrust laws?
Giving CC info to every Tom Dick and Harry is very irritating, not to mention security considerations. Think of all the times you've had to give them addresses and phone numbers, too. Even if the merchant doesn't rip you off, that's one more chance of your personal information being added to a list.
It would be nice if we could all have those nice checking account features that banks lavish on businesses, such as positive payments. A more ideal situation would mean giving a merchant a number unique to them and that transaction and the relevant account, any other purposes would be invalid. Of course, you probably couldn't do this because it would be easier to boil the oceans than to get the banking system to change.
Companies will prioritize who they want to lay off first. Making your job harder to be done by someone else will decrease your chances of being laid off if the product or support function for the company is needed. In that sense, job security does exist. To what extent is a different question.
PCs are disposable equipment, they are not durable goods by any stretch of the imagination.
Most businesses would rather minimize their expenses, especially IT, if they could. If I were the King of Gartner Research, I'd conduct a poll on how many business customers are irritated at the pace of upgrades. I bet quite a few.
Most of todays upgrades don't buy much ROI, but are done to prevent further expenses later, such as lack of support. What would I, as an ordinary consumer get from spending the $30 for 256MB SIMMS, other than a receipt for the expenditure? Could I better analyze my finances? Could I play games that I couldn't before? Could I then run a big behemoth ERP system for my household? I just don't see the *PRACTICAL* gain from this. It may be faster, but I had to spend money to do that. Also, given todays programming, the speed with the new memory and new XP OS might be the same as the speed of a computer with lesser memory and an older OS.
From an accounting standpoint, you're probably right. I'm sure computers go on the books as an immediate expense, instead of amortizing the cost over the life of the true useful life of the machine. That's because the industry has created a mindset of instantaneous obsolesence, for their own benefit--not consumers of any form. I would guess that to be the practical journal entry, but it may be different because of IRS depreciation rules.
Users want a fat O/S distribution.
Sure. More CD's to go through. More features, maybe. But how many would they not alreayd have? Also, if they knew the time it took to load the excessive bloat, would they still want it?
Of course if your idea of computing is to use a Dec Vt100 attached to the serial port of your 386 box, then go ahead. Just don't expect the rest of us to adapt because you choose to use a museum piece.
If I were playing games, or surfing the web, I'd need the absolute fastest largest machine. But if I wanted to get work done, the vt100/386 combo might be faster, because the programmer couldn't bog it down with worthless graphical, web, javascript bloat. *Note* The previous was an exageration.
Why can't MS, and much of the rest of the programming community, stop slowing down their code so that programs always run as slow as they did, regardless of the speed of the machine. Is that too hard to ask? Stop making the poor, and apparently wiser, HW engineers work much harder to keep up with horrible, space/time unconserving practices.
Re:This is not the traditional embedded market
on
Windows XP Embedded
·
· Score: 2, Informative
The self checkout market has many vendors. Having worked with one for Kroger, I can assure you that the back end at least, won't be Windows based. I'm reasonably certain the user GUI won't be either. The Kroger supermarket application runs on IBM's 4690 POS OS, which is based on a DOS derivative.
The 4690 OS much more light weight, designed for reliability and quality, and to a greater extent, upgradability. It's also been around for over 10 years. Stores have been known to run for two weeks on backup servers without even knowing it was switched over.
Also, different than Fat MS, I've been told to keep my code small so that the base Kroger supermarket app can fit within 1mb of ram. I'm not saying everything should be programmed this way, but the constraints put upon by *good* kiosks, cash registers, ATM's, and especially embedded devices are too heavy for something like XP.
Re:This is not the traditional embedded market
on
Windows XP Embedded
·
· Score: 1
I agree. I joke about this, but maybe it'd be more reliable, and cheaper to build a kiosk or a desk, staff it 24x7 with cheap labor and paper, and have human beings than to pay support costs for MS products. Is quality going down?
Re:This is not the traditional embedded market
on
Windows XP Embedded
·
· Score: 1
I admit to saying positive things about MS products and ideas in the past. However, after using W2000 last night, suffering the slowness that it brought on, I wonder who would use XP in even a PDA or museum kiosk. Browsing, and searching activities have slowed to the point that it's now faster to use paper and pencil. Let's hope Linux doesn't learn from that aspect of software misengineering.
I thought XML and SOAP were very tightly integrated with.NET. Also, keep in mind that MS is a decision maker on the XML board, so it's not like they're totally against it.
I think it takes courage to say something nice about MS on this site. Like a prior post said, if you use Windows, C# would be worth considering. It appears to provide more libraries, as part of the base offering, than C++, while offering additional protection and stability. Additionally, if I'm not mistaken, you get the GUI building tools which I've always found painfully easy to use. Contrast that with Sun, who doesn't have such features and makes you manually type in all the AWT/Swing stuff by hand. I'm sure there's supplemental products out there which might alleviate this, but it's still not as easy as having it built in.
Another point about the FS community bashing MS. If the oss community doesn't like C# because it's "closed" and made by MS, why doesn't it create it's own Java like language instead of relying on Sun? Seems to me that Sun still controls Java, hasn't truely made it an oss language, it has only made it available for free.
Btw, does anyone else not like the.NET name? It's petty, but how about something that doesn't sound internet related.
I really hate this mentallity. Though it is true that hardware is getting faster, that's no excuse for slow code. Doesn't anyone want to have "instantaneous" response times when they click on something? I, for one, don't want to see a spreadsheet on a website, where in I type something in, and have to click some button somewhere, and watch the comet circle around. Please don't make everything run in a browser! (Not that you explicitly suggested that, though).
People commonly say Linux is smaller than Windows, and can run on older hardware. The MS Windows way is to say, go ahead. Make the software slower, the computers will get faster!
Also, just think how much more efficient, and pleasant, computing would be if people did care about speed and footprint size.
Lastly, I've almost given up hope on the above. People continuously will want to make a program in 5 minutes that many people will use for a short time instead of spending more time developing quality, robust software that will be used for years.
I think the troll moderation is undeserved. The C++ vs Java comparison may be slanted, but at least has justifications to back up the opinion.
Before you think Java is fast enough, just run Ant on a large project. It's either slow because it's based on XML or slow because it's Java, or both. Yet, it is powerful, and has some elegance to it.
Also, to the Java crowd--Don't forget the original post said, C++. Convincing someone to use a different language and/or platform is something of a skill aside from programming, and might not be possible. This is a business issue, not a technical one.
Slashdot Mirror
All good suggestions. As I posted on a higher level, people focus too much on skins. Others I suggest:
* Make use of the Win Key! (Or some other replacement). It's really nice to have ONE keystroke that gets me almost everywhere I need to go under windows. Also, copy the Ctrl-Insert, Shift-Delete, combo's.
* Merge better with XFree project. Wouldn't it be great if I could simultaneously configure KDE and X because they share similar options?
* Somehow better integrate KDE with other parts of Linux so that similar devices don't have to be configured twice. On an earlier release, for instance, I had to define the backspace key both in X and under the console mode.
* Better proxy support, ideally for all relevant applications--configured in one place. I really hate dealing with firewalls, and this might be a good place to do battle with those things.
* Never make me edit a text file to configure a system setting. The less I have to know about what file to edit, where it is, and what the format is, the better and faster I'd be.
RealBoring
Can we all get off this theme/skin kick? Really, the whole point of the new release was better connectivity between applications. Better integration. People who worship command lines should focus more on how things work, not on how things look.
Since we can't expect every web person to consider good design, here's some future requirements for browsers:
* Provide an easy to reach on/off switch for irritants like large images animations and flash.
* If I have flash turned off, don't tell me so every time (IE does this)
* Allow me to turn off all pictures/animations/etc and click on the page to then selectively download just that one.
* Allow a total popup ban, without some third party tool.
* Don't allow the browser to become full screen, or let me turn this off.
* Always give me the minimize, restore and close buttons, never let some web bastard take them away from me.
* Let me specify the Java version I use.
* Give me multiple proxy options--our's a frequently down for whatever reason, and I don't want to dig deep into options to manually switch them. And for that matter, help me figure out what proxy is being used, since I can't find that out easily.
* Minimize the unknown information disclosed to the web server.
Maybe so, but they still should include some form of assembler. It doesn't much matter what platform, so long as they have a clue on what it really takes for the computer to do something. Too often, programmers write sloppy, inefficient, badly patterned code with their only justification being that CPU's keep getting faster, and memeory cheaper.
That would be hard to say since mainframe OS's are leased and priced by processor power, not a fixed fee.
Speed is probably priority 4 or 5. It's ease of use that's 8 or 9.
Which OS were you refering to then? z/OS?
I agree that I'd be really nice to cut down on the footprint size, and power usage. But I don't need the cute ruby, saphire, boxes with the rounded shapes. For me, I'd rather have this:
I guess I'm unique in this. I just want something boring, unfashionable, ugly, but useful and upgradeable.
On Microsoft:
MSDN used to be on CD's (possibly still true today, I'm not sure) which would have been far faster then any website. But then everything has to be web based right? But maybe it's just bad site design. I too was put on hold for close to an hour for tech support on MS Access, and I eventually gave up, called the premium line anyway, and actually got someone. He attempted to solve my problem, but mostly told me I couldn't do what I wanted because it didn't fit some paradigm of theirs.
A better example:
If you want good technical support, and you're willing to pay, go with IBM. Unlike MS, you can actually look up those cryptic error messages and not get 18 billion hits irrelevant hits. Things you need are organized into standard "books"-- references, guides, intro's, etc. When you need to quickly look something up, you know where to go, for specific answers not reading an entire RFC or gobligook, or ad laden website. I'm sure there are exceptions, though.
Also, support is something of a separate profession. People are actually measured for timeliness, quality, number of defects closed, etc. Also, fixes to fixes are heavily looked down upon.
You get this, because you pay for it, (quite a bit I'm told). But you don't have to wait for a plethora of people to have the problem to get it fixed. You also don't have to learn the internals to fix it like you would in an open source model.
One bad thing on Open Source support:
One time I wanted to turn off that annoying beep that the commandline mysql client has. When I posted my question to a mysql group, not IRC, No one could tell me how. One person even had the guts to tell me it's there for a reason, and I shouldn't try to circumvent it, but just put up with it! In my case, I just want to use Mysql, I don't want to get the source, find out where the beep comes from, and fix it myself. Sadly, I think that's the response one should expect to get when people don't get paid to help. I'm sure other's have had better results, and I'm happy they have, but I haven't seen the good results to be predictable like they would be with a support contract (not from Microsoft). If you had the money to pay for a support contract for a piece of OSS, then you'd have a different story.
In most OS's, you wouldn't want to do away with the CLI unless the GUI was absolutely perfect to your liking, and you never needed to do hard maintenance with it. I don't know what OS qualifies for this. I personally love to use vim, and find that to be one of my favorite *nix applications (of course I also use the NT version). Also, automatic administrative tasks would be much harder if we didn't have a good CLI.
Conversely, I also use Visual SlickEdit, which uses a lot of point and click, and makes navigating massive programs much easier. Sometimes I use the autocomplete feature. It's also nice to see method signatures popup in a tooltip. Or expand and collapse classes with a click. Also, before you dismiss the value of a GUI, think of the number of times you wanted to do file operations on files within a directory or on a set of files from a search result. It's a royal pain to have to scroll back up to see the directory contents, or worse, have to reissue the directory command.
I should point out, that although I do use GUI based tools, I don't use the mouse very often. I suspect that the most competent Windows users use a lot of the keyboard shortcuts. The GUI, in this situation, gives us a menu of possibilities so we don't have to memorize commands, but we can still use the keyboard and be fast about it.
Finally, CLI don't have to be difficult. The designers could use more verbose and intuitive commands and use a consistent, non symbol heavy syntax. My favorite example is the old VM/CMS command line. Getting a list of files: FILE. Checking your mail: READERLIST. And they could be shortened via synonyms or the first x characters of the command uniquely identifying it. Commands commonly used keyword parameters so at least you didn't need to know the order. Plus, some programs used a menu based interface (though it was text).
My conclusion: I already have a set of interfaces I'm used to using. Please, don't make other's switch because eventually someone is going to force that on me, too. (Ie, making Quicken more like a web browser instead of using the old MDI interface). Also, whatever it was that the researcher wanted to replace the desktop metaphor with sounded confusing when I read it, so maybe it wouldn't be as intuitive to use it.
I must be missing something. If you eliminate ACK packets, and reduce the amount of regular packets to "symbol" based packets, how can that possibly generate more traffic? This is an honest question, I'm not trying to insult anyone.
What would you define an OS to include. If it includes utilities such as a defragmenter that is also sold separately by another vendor, is that really so bad? In essence, how much of Windows, the OS itself and packaging would you remove to make it not violate antitrust laws?
It would be nice if we could all have those nice checking account features that banks lavish on businesses, such as positive payments. A more ideal situation would mean giving a merchant a number unique to them and that transaction and the relevant account, any other purposes would be invalid. Of course, you probably couldn't do this because it would be easier to boil the oceans than to get the banking system to change.
Companies will prioritize who they want to lay off first. Making your job harder to be done by someone else will decrease your chances of being laid off if the product or support function for the company is needed. In that sense, job security does exist. To what extent is a different question.
Most businesses would rather minimize their expenses, especially IT, if they could. If I were the King of Gartner Research, I'd conduct a poll on how many business customers are irritated at the pace of upgrades. I bet quite a few.
Most of todays upgrades don't buy much ROI, but are done to prevent further expenses later, such as lack of support. What would I, as an ordinary consumer get from spending the $30 for 256MB SIMMS, other than a receipt for the expenditure? Could I better analyze my finances? Could I play games that I couldn't before? Could I then run a big behemoth ERP system for my household? I just don't see the *PRACTICAL* gain from this. It may be faster, but I had to spend money to do that. Also, given todays programming, the speed with the new memory and new XP OS might be the same as the speed of a computer with lesser memory and an older OS.
From an accounting standpoint, you're probably right. I'm sure computers go on the books as an immediate expense, instead of amortizing the cost over the life of the true useful life of the machine. That's because the industry has created a mindset of instantaneous obsolesence, for their own benefit--not consumers of any form. I would guess that to be the practical journal entry, but it may be different because of IRS depreciation rules.
Sure. More CD's to go through. More features, maybe. But how many would they not alreayd have? Also, if they knew the time it took to load the excessive bloat, would they still want it?
If I were playing games, or surfing the web, I'd need the absolute fastest largest machine. But if I wanted to get work done, the vt100/386 combo might be faster, because the programmer couldn't bog it down with worthless graphical, web, javascript bloat. *Note* The previous was an exageration.
Why can't MS, and much of the rest of the programming community, stop slowing down their code so that programs always run as slow as they did, regardless of the speed of the machine. Is that too hard to ask? Stop making the poor, and apparently wiser, HW engineers work much harder to keep up with horrible, space/time unconserving practices.
The 4690 OS much more light weight, designed for reliability and quality, and to a greater extent, upgradability. It's also been around for over 10 years. Stores have been known to run for two weeks on backup servers without even knowing it was switched over.
Also, different than Fat MS, I've been told to keep my code small so that the base Kroger supermarket app can fit within 1mb of ram. I'm not saying everything should be programmed this way, but the constraints put upon by *good* kiosks, cash registers, ATM's, and especially embedded devices are too heavy for something like XP.
I agree. I joke about this, but maybe it'd be more reliable, and cheaper to build a kiosk or a desk, staff it 24x7 with cheap labor and paper, and have human beings than to pay support costs for MS products. Is quality going down?
I admit to saying positive things about MS products and ideas in the past. However, after using W2000 last night, suffering the slowness that it brought on, I wonder who would use XP in even a PDA or museum kiosk. Browsing, and searching activities have slowed to the point that it's now faster to use paper and pencil. Let's hope Linux doesn't learn from that aspect of software misengineering.
Are you sure the mainframe has the capacity to run XP? :)
I didn't realize police records were protected. Not that it doesn't make sense. I only wish I could do that.
I thought XML and SOAP were very tightly integrated with .NET. Also, keep in mind that MS is a decision maker on the XML board, so it's not like they're totally against it.
I think it takes courage to say something nice about MS on this site. Like a prior post said, if you use Windows, C# would be worth considering. It appears to provide more libraries, as part of the base offering, than C++, while offering additional protection and stability. Additionally, if I'm not mistaken, you get the GUI building tools which I've always found painfully easy to use. Contrast that with Sun, who doesn't have such features and makes you manually type in all the AWT/Swing stuff by hand. I'm sure there's supplemental products out there which might alleviate this, but it's still not as easy as having it built in.
Another point about the FS community bashing MS. If the oss community doesn't like C# because it's "closed" and made by MS, why doesn't it create it's own Java like language instead of relying on Sun? Seems to me that Sun still controls Java, hasn't truely made it an oss language, it has only made it available for free.
Btw, does anyone else not like the .NET name? It's petty, but how about something that doesn't sound internet related.
I really hate this mentallity. Though it is true that hardware is getting faster, that's no excuse for slow code. Doesn't anyone want to have "instantaneous" response times when they click on something? I, for one, don't want to see a spreadsheet on a website, where in I type something in, and have to click some button somewhere, and watch the comet circle around. Please don't make everything run in a browser! (Not that you explicitly suggested that, though).
People commonly say Linux is smaller than Windows, and can run on older hardware. The MS Windows way is to say, go ahead. Make the software slower, the computers will get faster!
Also, just think how much more efficient, and pleasant, computing would be if people did care about speed and footprint size.
Lastly, I've almost given up hope on the above. People continuously will want to make a program in 5 minutes that many people will use for a short time instead of spending more time developing quality, robust software that will be used for years.
I think the troll moderation is undeserved. The C++ vs Java comparison may be slanted, but at least has justifications to back up the opinion.
Before you think Java is fast enough, just run Ant on a large project. It's either slow because it's based on XML or slow because it's Java, or both. Yet, it is powerful, and has some elegance to it.
Also, to the Java crowd--Don't forget the original post said, C++. Convincing someone to use a different language and/or platform is something of a skill aside from programming, and might not be possible. This is a business issue, not a technical one.