They are working on it but nobody reports issues. Telling Slashdot "X11 from Apple sux, it can't use services" is not the way to go. You say "X11.org can't use Services, I have an idea how to implement them". It is not like Cocoa, it is completely open source.
I am not a developer of any kind, I try my best as end user to report issues I spot.
It is not a bad X11 (as they moved to x.org) , in fact Apple made possible and lot easier for X11 apps run in average end user desktop by integrating it to launchd subsystem. If you type "xmms" from command line, X11 launches in rootless window and opens xmms. It is that easy. I use Kopete as my instant messenger since I installed Leopard (thanks to finkproject.org ) and various tools like Koffice.
Real issue with X11 apps could be the thing that most of them are coded with Linux in mind. Those Fink and Macports guys spare a lot of time to make them compile in a true Unix/NeXT environment on a OS which even openly warns its own core tools like Finder.app NOT to use depreciated functionality (from system.log). That is how Apple expects people to code on the OS X. They warn politely first, a bit serious later and it basically refuses to launch even with a informative crash.
If I was Adobe, I would be very mad of tools which are native OS X apps (which uses latest technologies) and communicate with non GUI Unix tools (like imagemagick) which can be run as 64bit since Tiger 10.4.0. Add the Apple CoreImage to the program, you have something very powerful in hand.
They weren't seeing imagemagick or Core Image as threat until some gifted, real OS X developers started to take advantage of them. If Adobe coded their application in same manner (which Apple expected them to do), they could ship 64bit Photoshop in Tiger 10.4.0 ages.
If I was developing them, I would release a 64bit version tomorrow right while people argue about 64bit Photoshop. It could be simple as a single click in XCode 3.
Isn't Adobe making sure that they can't release a Linux/OpenStep based version by ignoring Cocoa BTW? Wouldn't it be lot simpler to port to Linux/OpenStep from Cocoa?
If we speak about Carbon, MS could be one of large companies to blame. Apple got sick of companies like Microsoft still pushing Carbon apps to users completely ignoring Cocoa and they have taken some measures against them on Leopard.
In fact, Microsoft got a surprise when XCode 3 refused to compile (and ship?) MS Office without them fixing the issues at their code first. http://blogs.msdn.com/macmojo/archive/2007/10/29/there-s-a-new-cat-in-town.aspx "When I led our transition from CodeWarrior to Xcode back in late 2005, I noted just how picky the tools were. I'm pleased to note that the gcc toolset is even pickier now, and has helped us find and fix a few bugs that Xcode 2.4 missed."
translation: It refused to compile:)
Apple's philosophy is really different from MS. They say "We are moving fast, if you keep up with us, that is fine. If you don't? Your app icon will bounce and it won't launch".
Adobe Photoshop could be huge, really huge but there is another huge thing which will ship 64bit Leopard version completely moving to Cocoa. Trolltech Qt 4. They had same issue, they heard the Carbon 64bit issue from WWDC. They didn't use it as excuse (like Adobe) and I am sure they are working on 64bit Qt 4 right now. Add Microsoft too. Seriously, a XCode 3 compiling, Cocoa Microsoft Office 08 is a huge thing. Besides jokes, we should congratulate them for not using it as excuse.
As they are unofficial monopoly, nobody can dare to shutdown them without breaking entire country. That is how they dare to do such things. It is similar in lots of countries.
"A lot of the stuff they are working on is pretty cool, especially Mono."
I wonder how many people/IT chiefs rejected to use Novell solutions and Suse since they were seriously irritated by the cool Mono developer and his actions like "installing.NET to iPhone", "OOXML is superb", the organization he founded still can't say a word about horrible IT/ISO scandal, their tactics of "Use our products or MS may sue you in future".
Novell was a great company while they were actually competing with Microsoft. Now, for many, they are only MS Trojans.
Acid 3 pass reminds me the safe way of developing. Put a perfectly running application first and optimize on it. It is just a number otherwise.
I prefer a complete application and let it get optimized on it. Opera 9.5 will have to get optimized since they have to sell same html rendering code to me, running Symbian S60 V3 Nokia E65 which has 24 MB of free RAM.;) It runs Opera 8.65 currently even with antivirus/security/antispam (Kaspersky Mobile Security) overhead and a all feature instant messenger/VOIP client (Fring).
I was afraid that Opera guys would get into that pointless "speed race" of 2007/2008 season and forget their real focus. I have nothing against optimization, you can see my various questions about why some parts of browsers aren't using Altivec (PPC) code over the web. I am just saying, a good core first.
If they could push Youtube to use a real UDP/RTSP bandwidth switching, real/quicktime/vlc plugin, even as option...
Of course Youtube must keep using 1994's "embed a huge file and let browser play while downloading if it can" technology (!). If I was Adobe, I would spare thousands of developers to fix that fake media plugin and even buy the bandwidth autoswitching patents from Real Networks (which are free to GNU/OSF software).
If they weren't afraid of Google Inc. they would have already started to cap flash/flv packets. It doesn't mean they don't have such plans. I bet ISPs Worldwide already staring at their Youtube bandwidth waste.
Another option would be a P2P based video streaming plugin for Flash but we are speaking about Comcast here;)
DOCSIS 3.0 has very sophisticated Quality Of Service (can read as unbreakable capping) technology built in. If you think it will be easy to hack, believe me nothing on DOCSIS is easy. Saying as a person who had to throw away his Motorola SB3100 because ISP rejected to update its firmware.
I tried 2 weeks before giving up and it was supposed to be easier than others since (archaic) modem had built in software upgrade option from DOCSIS 1.0 to 1.1.
We are speaking about a ISP sent RESET signal to their customers even if they download Linux ISO images or licensed/paid movies.
OOXML is more than official standards of HTML, a lot more. Also even if it was just HTML, I wouldn't trust word of company who undermined the official HTML standards for years and just recently promised to support them in full right after Opera Software sued them.
Apologies for not being so neutral about a framework (!) since I had to run Virtual PC, a freaking emulator to run a program (idiot Nokia, hope iPhone kills you) written in that ECMA standard, multiplatform (!) crap.
Java exists and runs on my native OS X which does generate heavily optimised PPC Altivec/JIT ASM code realtime. That is a real framework,.NET is a bulky windows DLL. Nothing else.
The only and strictly single way of opening a OOXML document in its pixel/inch perfect form with every detail is having $400 MS Office '08 which even OS X version has some problems.
They voted that "thing" to be a international standard. It is basic as that.
If a thing runs on only one OS/platform (Win32/64 on x86) and it is interpreting things on exact same machines which runs C programs happily, I would wonder the reasons of shipping such a useless product. It must have some other reasons rather than the proposed one.
Look at Apple, they are shipping their frameworks one by one to Windows and they never claimed they code multiplatform standard. They say "Quicktime is multiplatform" and you see the result, sometimes Windows Quicktime works better and more compatible than the Quicktime on their own OS. They silently allow open source products to ship Quicktime reading/writing apps which sometimes shows their own product as a joke.
I say, if there is an example of a true multiplatform, compile once and run anywhere product, it is Java. I give only 2 software names, "Opera Mini" and "Azureus". Single JAR, runs anywhere and more advanced than pure C apps. If J2EE wasn't true "write once run anywhere", mainframe guys wouldn't even touch it too.
A standards body is non existent if people/governments and large corporations openly accuses them or suspects fraud.
They would gather and setup ISO 2.0. If you think about JPEG being an ISO standard, you can imagine how important ISO is. Wonder if they will approve "HD Photo" from Microsoft as a ISO standard too?
With that decision, they are open to every kind of accusation. A company can remind "OOXML" when a completely irrelevant standard passes which would be in favour of a large corporation.
Companies should ask themselves "What if Microsoft goes chapter 11 in matter of a decade" and decide their formats based on that. All computers I owned in 1980s were made by huge, untouchable large companies which nobody could even imagine their fall. It happened.
It serves great to teach non technical home users about picking right ISP, using SSL all places.
You can lecture them for hours and they will still use horribly insecure things. You fire up Wireshark with default settings and tell their ISP or that Coffee house (with wireless) admin "can run it". It is like shock theraphy. When they figure the amount of data their ISP can trace about them, they may find a better and trusted one too.
Investigation should have started right after they gathered ISO support for.NET. Sold out geeks kept claiming.NET is a standard and while they are stuck on version 1.0 of standard with their "mono",.NET 3.0 Apps are all over the place..NET became Dolby Digital EX while they are stuck in Mono.
See the story:
An article up at Intellectual Property Watch claims they have obtained a document (PDF)
See the article linked is "PDF"? Why? It is supported on everything down to Symbian S60 handsets and any open source software can support it. People can even race with vendors "reader" software making better ones. That is a real standard which won its place without dirty tricks.
I bet usual suspects like Novell and their mighty Mono/Silverlight innovator Icaza will come up with a thing that supports it to some extent, advertise it and MS will use it as a proof.
Last question: Did gnome people openly critised this decision? On their website?
April 1 could be the end of ISO. Once you lose credibility, you don't get it back. It is not a April 1 joke either. You can even feel that one of the biggest IT scandals is waiting and this time it is not poor open source geeks anymore, it is IBM/Sun and GNU/BSD and various World governments especially those very rich ones who can even say "no" to EU. Don't forget the militaries either.
In worst case, a competitor or a mad worker may easily install a software from a company which is known to hate piracy, call BSA and create a major issue. It is a risk really. There is also $$$ involved since companies give money to reporters.
Same kind of pointless naming war leads to the author working at CNET claim "Firewire" is dead. He probably got Firewire and he is not aware of it since Sony names it i-Link and others use IEEE-1394! name.
I don't get what planet that guy lives, certainly not Europe.
Every single SD (non HD) device has SCART connector especially set top boxes. High End always used Component cable, some ultra small DVD players won't have SCART because of its space issues. It is not a invented thing, it is designed to make it easy for average customer to plug their devices to TV and it still does it job perfectly even today.
I think guy tries to be sarcastic and "The Register" like writing the article (bitching about French) but SCART isn't the right target. Only thing to replace SCART is HDMI which is digital and has SCART-like features (Chaining etc.) along with evil DRM.
A very high end proactive/heuristic (really heuristic!) antivirus/security suite may figure such attack but naturally they are only available on Windows for obvious reasons.
Win2k3 "server" without any updates may run fine since it is used in enterprise, running only Oracle server for example. I saw actual Win XP Home edition SP2 just sitting on a 2 client Wireless network today, some genius disabled its "automatic updates" because it takes too long for machine to turn off (ms geniuses new methods) and same XP suicidal guy also disabled firewall.
You should have seen the circus on that machine. It took 3 hours to clean viruses/spyware/rootkits.
BTW Symantec's Mac Antivirus department seems to wake up recently and they promise "exploit" protection but I don't think it is heuristic in way of Windows ultra paranoid stuff like Kaspersky or old eSafe Desktop.
Lets not forget the ultimate practical inventions on SCART (which is still used!) like sending TV/Display "Switch to me" signal. (I guess over pin18) It is amazing that people designed HDMI missed it.
In Europe, when you turn on a set top box, it will send a signal to TV from a special pin saying "Switch to me" and your TV will automatically switch to the device. If it is high end TV, it may ignore with a setting though. Some devices also send "Release my channel" and if your TV is wise enough, it will go back to last input source (or tuner). At least my cheap DVB-S receiver does it.
They design a digital interface in 2000s and forget to put such thing in spec. HDMI could get much more popular if people didn't to click a button 4-5 times to switch to a satellite.
Another guy mentioned: You design a thing which should replace SCART, promise people it is not just evil DRM, it is for ease of use and you still make it "Can be plugged one way only". At some houses, replacing a broken HDMI cable may need 2-3 guys, to handle the display.
CNET is a IT oriented site, they have hard time to understand the TV World and why TV guys always "Stick with working thing". SCART is a thing from 1977, it will be there until EBU decides it is obsolete. TV doesn't work like computers, you can't fool around with standards unless there is absolute need for change. Whoever designed SCART and made it patent free (or cheap) with such scalability deserves a award for it.
Slashdot Mirror
They are working on it but nobody reports issues. Telling Slashdot "X11 from Apple sux, it can't use services" is not the way to go. You say "X11.org can't use Services, I have an idea how to implement them". It is not like Cocoa, it is completely open source.
I am not a developer of any kind, I try my best as end user to report issues I spot.
http://trac.macosforge.org/projects/xquartz .
It is not a bad X11 (as they moved to x.org) , in fact Apple made possible and lot easier for X11 apps run in average end user desktop by integrating it to launchd subsystem. If you type "xmms" from command line, X11 launches in rootless window and opens xmms. It is that easy. I use Kopete as my instant messenger since I installed Leopard (thanks to finkproject.org ) and various tools like Koffice.
Real issue with X11 apps could be the thing that most of them are coded with Linux in mind. Those Fink and Macports guys spare a lot of time to make them compile in a true Unix/NeXT environment on a OS which even openly warns its own core tools like Finder.app NOT to use depreciated functionality (from system.log). That is how Apple expects people to code on the OS X. They warn politely first, a bit serious later and it basically refuses to launch even with a informative crash.
If I was Adobe, I would be very mad of tools which are native OS X apps (which uses latest technologies) and communicate with non GUI Unix tools (like imagemagick) which can be run as 64bit since Tiger 10.4.0. Add the Apple CoreImage to the program, you have something very powerful in hand.
http://www.versiontracker.com/dyn/moreinfo/macosx/26426 (iMaginator)
http://www.versiontracker.com/dyn/moreinfo/macosx/30452 (ChocoFlop)
http://www.versiontracker.com/dyn/moreinfo/macosx/33411 (Pixelmator which uses imagemagick)
They weren't seeing imagemagick or Core Image as threat until some gifted, real OS X developers started to take advantage of them. If Adobe coded their application in same manner (which Apple expected them to do), they could ship 64bit Photoshop in Tiger 10.4.0 ages.
If I was developing them, I would release a 64bit version tomorrow right while people argue about 64bit Photoshop. It could be simple as a single click in XCode 3.
Isn't Adobe making sure that they can't release a Linux/OpenStep based version by ignoring Cocoa BTW? Wouldn't it be lot simpler to port to Linux/OpenStep from Cocoa?
If we speak about Carbon, MS could be one of large companies to blame. Apple got sick of companies like Microsoft still pushing Carbon apps to users completely ignoring Cocoa and they have taken some measures against them on Leopard.
:)
In fact, Microsoft got a surprise when XCode 3 refused to compile (and ship?) MS Office without them fixing the issues at their code first.
http://blogs.msdn.com/macmojo/archive/2007/10/29/there-s-a-new-cat-in-town.aspx
"When I led our transition from CodeWarrior to Xcode back in late 2005, I noted just how picky the tools were. I'm pleased to note that the gcc toolset is even pickier now, and has helped us find and fix a few bugs that Xcode 2.4 missed."
translation: It refused to compile
Apple's philosophy is really different from MS. They say "We are moving fast, if you keep up with us, that is fine. If you don't? Your app icon will bounce and it won't launch".
Adobe Photoshop could be huge, really huge but there is another huge thing which will ship 64bit Leopard version completely moving to Cocoa. Trolltech Qt 4. They had same issue, they heard the Carbon 64bit issue from WWDC. They didn't use it as excuse (like Adobe) and I am sure they are working on 64bit Qt 4 right now. Add Microsoft too. Seriously, a XCode 3 compiling, Cocoa Microsoft Office 08 is a huge thing. Besides jokes, we should congratulate them for not using it as excuse.
As they are unofficial monopoly, nobody can dare to shutdown them without breaking entire country. That is how they dare to do such things. It is similar in lots of countries.
"A lot of the stuff they are working on is pretty cool, especially Mono."
.NET to iPhone", "OOXML is superb", the organization he founded still can't say a word about horrible IT/ISO scandal, their tactics of "Use our products or MS may sue you in future".
I wonder how many people/IT chiefs rejected to use Novell solutions and Suse since they were seriously irritated by the cool Mono developer and his actions like "installing
Novell was a great company while they were actually competing with Microsoft. Now, for many, they are only MS Trojans.
Acid 3 pass reminds me the safe way of developing. Put a perfectly running application first and optimize on it. It is just a number otherwise.
;) It runs Opera 8.65 currently even with antivirus/security/antispam (Kaspersky Mobile Security) overhead and a all feature instant messenger/VOIP client (Fring).
I prefer a complete application and let it get optimized on it. Opera 9.5 will have to get optimized since they have to sell same html rendering code to me, running Symbian S60 V3 Nokia E65 which has 24 MB of free RAM.
I was afraid that Opera guys would get into that pointless "speed race" of 2007/2008 season and forget their real focus. I have nothing against optimization, you can see my various questions about why some parts of browsers aren't using Altivec (PPC) code over the web. I am just saying, a good core first.
If they could push Youtube to use a real UDP/RTSP bandwidth switching, real/quicktime/vlc plugin, even as option...
;)
Of course Youtube must keep using 1994's "embed a huge file and let browser play while downloading if it can" technology (!). If I was Adobe, I would spare thousands of developers to fix that fake media plugin and even buy the bandwidth autoswitching patents from Real Networks (which are free to GNU/OSF software).
If they weren't afraid of Google Inc. they would have already started to cap flash/flv packets. It doesn't mean they don't have such plans. I bet ISPs Worldwide already staring at their Youtube bandwidth waste.
Another option would be a P2P based video streaming plugin for Flash but we are speaking about Comcast here
As we are supposed to speak about Comcast, I remembered Azureus bittorrent client. ;)
It has "Use fancy tabs" setting in advanced mode. Slashdot may have "Use Fancy buttons" which thousands of geeks will disable.
DOCSIS 3.0 has very sophisticated Quality Of Service (can read as unbreakable capping) technology built in. If you think it will be easy to hack, believe me nothing on DOCSIS is easy. Saying as a person who had to throw away his Motorola SB3100 because ISP rejected to update its firmware.
I tried 2 weeks before giving up and it was supposed to be easier than others since (archaic) modem had built in software upgrade option from DOCSIS 1.0 to 1.1.
We are speaking about a ISP sent RESET signal to their customers even if they download Linux ISO images or licensed/paid movies.
OOXML is more than official standards of HTML, a lot more. Also even if it was just HTML, I wouldn't trust word of company who undermined the official HTML standards for years and just recently promised to support them in full right after Opera Software sued them.
Apologies for not being so neutral about a framework (!) since I had to run Virtual PC, a freaking emulator to run a program (idiot Nokia, hope iPhone kills you) written in that ECMA standard, multiplatform (!) crap.
.NET is a bulky windows DLL. Nothing else.
Java exists and runs on my native OS X which does generate heavily optimised PPC Altivec/JIT ASM code realtime. That is a real framework,
The only and strictly single way of opening a OOXML document in its pixel/inch perfect form with every detail is having $400 MS Office '08 which even OS X version has some problems.
They voted that "thing" to be a international standard. It is basic as that.
If a thing runs on only one OS/platform (Win32/64 on x86) and it is interpreting things on exact same machines which runs C programs happily, I would wonder the reasons of shipping such a useless product. It must have some other reasons rather than the proposed one.
Look at Apple, they are shipping their frameworks one by one to Windows and they never claimed they code multiplatform standard. They say "Quicktime is multiplatform" and you see the result, sometimes Windows Quicktime works better and more compatible than the Quicktime on their own OS. They silently allow open source products to ship Quicktime reading/writing apps which sometimes shows their own product as a joke.
I say, if there is an example of a true multiplatform, compile once and run anywhere product, it is Java. I give only 2 software names, "Opera Mini" and "Azureus". Single JAR, runs anywhere and more advanced than pure C apps. If J2EE wasn't true "write once run anywhere", mainframe guys wouldn't even touch it too.
A standards body is non existent if people/governments and large corporations openly accuses them or suspects fraud.
They would gather and setup ISO 2.0. If you think about JPEG being an ISO standard, you can imagine how important ISO is. Wonder if they will approve "HD Photo" from Microsoft as a ISO standard too?
With that decision, they are open to every kind of accusation. A company can remind "OOXML" when a completely irrelevant standard passes which would be in favour of a large corporation.
Companies should ask themselves "What if Microsoft goes chapter 11 in matter of a decade" and decide their formats based on that. All computers I owned in 1980s were made by huge, untouchable large companies which nobody could even imagine their fall. It happened.
It serves great to teach non technical home users about picking right ISP, using SSL all places.
You can lecture them for hours and they will still use horribly insecure things. You fire up Wireshark with default settings and tell their ISP or that Coffee house (with wireless) admin "can run it". It is like shock theraphy. When they figure the amount of data their ISP can trace about them, they may find a better and trusted one too.
It seems the non biased opinions about "How great .NET 3.0 is?" to slashdot should be reviewed too.
.gov , .mil, .edu and .com which will certainly investigate this Apri fools joke, it could be the last ever mistake of MS.
I just had a 45 minute installation session for ".NET 3.0" on my MS Virtual PC 7 while the "bulky" "bloated" Java 6 for Windows takes 5 minutes.
Now I started to wonder who are those ".NET 3.0 is great" guys and by any chance, they got some IP in Bangalore.
Considering the number of
Investigation should have started right after they gathered ISO support for .NET. Sold out geeks kept claiming .NET is a standard and while they are stuck on version 1.0 of standard with their "mono", .NET 3.0 Apps are all over the place. .NET became Dolby Digital EX while they are stuck in Mono.
See the story:
An article up at Intellectual Property Watch claims they have obtained a document (PDF)
See the article linked is "PDF"? Why? It is supported on everything down to Symbian S60 handsets and any open source software can support it. People can even race with vendors "reader" software making better ones. That is a real standard which won its place without dirty tricks.
I bet usual suspects like Novell and their mighty Mono/Silverlight innovator Icaza will come up with a thing that supports it to some extent, advertise it and MS will use it as a proof.
Last question: Did gnome people openly critised this decision? On their website?
April 1 could be the end of ISO. Once you lose credibility, you don't get it back. It is not a April 1 joke either. You can even feel that one of the biggest IT scandals is waiting and this time it is not poor open source geeks anymore, it is IBM/Sun and GNU/BSD and various World governments especially those very rich ones who can even say "no" to EU. Don't forget the militaries either.
It seems the most practical way of preventing such issues is not allowing users to install anything. There must be some policy against it.
It is documented at that page http://acumenis.com/articles/2007/09/foiling-malware-with-software-restriction-policies which is focused on malware but can be applied to anything, even home machines.
In worst case, a competitor or a mad worker may easily install a software from a company which is known to hate piracy, call BSA and create a major issue. It is a risk really. There is also $$$ involved since companies give money to reporters.
There are reporting software for system admins to figure what has been licensed or not.
For example:
http://www.cidiscovery.com/
There must be lots of more options but funny is, they seem to be one of most pirated software on planet so I couldn't find more.
Same kind of pointless naming war leads to the author working at CNET claim "Firewire" is dead. He probably got Firewire and he is not aware of it since Sony names it i-Link and others use IEEE-1394! name.
Mozilla servers or Google servers? Think again if you are in conspiracy mood ;)
I don't get what planet that guy lives, certainly not Europe.
Every single SD (non HD) device has SCART connector especially set top boxes. High End always used Component cable, some ultra small DVD players won't have SCART because of its space issues. It is not a invented thing, it is designed to make it easy for average customer to plug their devices to TV and it still does it job perfectly even today.
I think guy tries to be sarcastic and "The Register" like writing the article (bitching about French) but SCART isn't the right target. Only thing to replace SCART is HDMI which is digital and has SCART-like features (Chaining etc.) along with evil DRM.
A very high end proactive/heuristic (really heuristic!) antivirus/security suite may figure such attack but naturally they are only available on Windows for obvious reasons.
Win2k3 "server" without any updates may run fine since it is used in enterprise, running only Oracle server for example. I saw actual Win XP Home edition SP2 just sitting on a 2 client Wireless network today, some genius disabled its "automatic updates" because it takes too long for machine to turn off (ms geniuses new methods) and same XP suicidal guy also disabled firewall.
You should have seen the circus on that machine. It took 3 hours to clean viruses/spyware/rootkits.
BTW Symantec's Mac Antivirus department seems to wake up recently and they promise "exploit" protection but I don't think it is heuristic in way of Windows ultra paranoid stuff like Kaspersky or old eSafe Desktop.
Lets not forget the ultimate practical inventions on SCART (which is still used!) like sending TV/Display "Switch to me" signal. (I guess over pin18) It is amazing that people designed HDMI missed it.
In Europe, when you turn on a set top box, it will send a signal to TV from a special pin saying "Switch to me" and your TV will automatically switch to the device. If it is high end TV, it may ignore with a setting though. Some devices also send "Release my channel" and if your TV is wise enough, it will go back to last input source (or tuner). At least my cheap DVB-S receiver does it.
They design a digital interface in 2000s and forget to put such thing in spec. HDMI could get much more popular if people didn't to click a button 4-5 times to switch to a satellite.
Another guy mentioned: You design a thing which should replace SCART, promise people it is not just evil DRM, it is for ease of use and you still make it "Can be plugged one way only". At some houses, replacing a broken HDMI cable may need 2-3 guys, to handle the display.
CNET is a IT oriented site, they have hard time to understand the TV World and why TV guys always "Stick with working thing". SCART is a thing from 1977, it will be there until EBU decides it is obsolete. TV doesn't work like computers, you can't fool around with standards unless there is absolute need for change. Whoever designed SCART and made it patent free (or cheap) with such scalability deserves a award for it.