AboutWelcome to Free Software Daily (FSD). FSD is a hub for news and articles by and for the free and open source community. FSD is a community driven site where members of the community submit and vote for the stories that they think are important and interesting to them. Click the "About" link to read more...
Oh, my god! Someone did find the honeypot of interoperability!!!
Quote from the article:
"Microsoft has thousands of developers working in lockstep on getting all of its applications to work together. There's never a worry that Microsoft Office will function effortlessly with Microsoft Exchange, Microsoft SharePoint, Microsoft Dynamics and so forth. Can the same be said across open-source applications? I would say the answer to that question is: not yet.
This challenge needs to be addressed by the open-source business community — and fast. While one of the greatest strengths of open-source development is its ability to rapidly produce a variety of features and applications, it becomes a weakness if these components cannot easily work together. To compete against proprietary vendors interoperability is vital. Without it, open source will fail.
This is not a challenge that can be addressed with point-to-point partnerships among individual commercial open-source vendors. Such partnerships may deliver a better point solution, but they barely scratch the surface of the broader interoperability problem. Corporations consume many point solutions and need to stitch them together in a variety of ways. Moreover, they frequently need to integrate with existing infrastructure and applications, many of which are proprietary. [...] Solving this problem through traditional bilateral partnering doesn't scale. There must be a better way."
Unfortunately, guy only talks about interoperability of commercial open source enterprise products, no word about small free software projects or about integration between desktop applications. What? You thought desktop is perfectly integrated? Think again!
I would like to be able to set up Firefox to use the KDE password manager, but it seems I'm out of luck. I would like that KMail and Thunderbird and Evolution shared a single set of settings, and allowed me to quickly switch to a new mail client, that just works right after the installation. There should be no need to import/ copy files from one settings folder to another.
Also what about a common torrent download engine, for the desktop?
Switching from Azureus to Miro and then to KTorrent should not diminish my reputation, as it does now, and I should not be required to redownload the files.
"I would like that KMail and Thunderbird and Evolution shared a single set of settings, and allowed me to quickly switch to a new mail client, that just works right after the installation."
As I am sure many would like that Outlook/Lotus et al would share common settings!
Sorry, but what planet are you on?
Interoperability is very unlikely to include sharing common settings in the way you desire. It's more likely to be about sharing data between applications.
As for Firefox/KWallet integration: I admit it would be nice for some (personally I use my head to remember passwords - it's more portable). It's not impossible but it may be impractical.
KWallet has a published API and applications that integrate with it adhere to that API. I've not seen many non-KDE applications that do so. Mozilla could probably integrate with it but would they want to stop at KDE support for this? Firefox is deliberately platform-independent ( granted it does use the host-system's dialogs ). Besides, I'm not so certain than the other platforms it can run on have an equivalent to KWallet. I don't recall Windows, OS X or GNOME having a central app which permits password storage across _all_ (inclusing non-native) applications.
I think this is pretty much FUD. Granted it's not from Microsoft or another proprietary company but it does take the approach of "we can't money from this unless we change the model" - which has never really been the point of free software in particular.
FTA: "Microsoft has thousands of developers working in lockstep on getting all of its applications to work together. There's never a worry that Microsoft Office will function effortlessly with Microsoft Exchange, Microsoft SharePoint, Microsoft Dynamics and so forth. Can the same be said across open-source applications? I would say the answer to that question is: not yet."
This is basically his premise: "Microsoft's products work together perfectly-so should OpenSource ones".
I have two issues with this:
1. MSOffice and Exchange do not always function effortlessly together. Take a look at Experts Exchange or similar and see how many issues there are around interoperability betwene MS products.
2. Open source products are not made by the same company so to expect the same standard (even though MS doesn't achieve it) is unfair. Do MS SQLServer and Oracle integrate seamlessly? Do MS Office and say Adobe Acrobat (not reader)? Then why should we demand that Open Source applications do?
Aside from that adherence to open standards by all would pretty much guarantee interoperability - certainly on a data front.
bogdanbiv
16 years 10 weeks 3 days 11 hours ago
Open Source Interoperability is the honey pot
Oh, my god! Someone did find the honeypot of interoperability!!!
Quote from the article:
"Microsoft has thousands of developers working in lockstep on getting all of its applications to work together. There's never a worry that Microsoft Office will function effortlessly with Microsoft Exchange, Microsoft SharePoint, Microsoft Dynamics and so forth. Can the same be said across open-source applications? I would say the answer to that question is: not yet.
This challenge needs to be addressed by the open-source business community — and fast. While one of the greatest strengths of open-source development is its ability to rapidly produce a variety of features and applications, it becomes a weakness if these components cannot easily work together. To compete against proprietary vendors interoperability is vital. Without it, open source will fail.
This is not a challenge that can be addressed with point-to-point partnerships among individual commercial open-source vendors. Such partnerships may deliver a better point solution, but they barely scratch the surface of the broader interoperability problem. Corporations consume many point solutions and need to stitch them together in a variety of ways. Moreover, they frequently need to integrate with existing infrastructure and applications, many of which are proprietary. [...] Solving this problem through traditional bilateral partnering doesn't scale. There must be a better way."
Unfortunately, guy only talks about interoperability of commercial open source enterprise products, no word about small free software projects or about integration between desktop applications. What? You thought desktop is perfectly integrated? Think again!
I would like to be able to set up Firefox to use the KDE password manager, but it seems I'm out of luck. I would like that KMail and Thunderbird and Evolution shared a single set of settings, and allowed me to quickly switch to a new mail client, that just works right after the installation. There should be no need to import/ copy files from one settings folder to another.
Also what about a common torrent download engine, for the desktop?
Switching from Azureus to Miro and then to KTorrent should not diminish my reputation, as it does now, and I should not be required to redownload the files.
crimperman
16 years 10 weeks 1 day 55 min ago
cloud cuckoo land
"I would like that KMail and Thunderbird and Evolution shared a single set of settings, and allowed me to quickly switch to a new mail client, that just works right after the installation."
As I am sure many would like that Outlook/Lotus et al would share common settings!
Sorry, but what planet are you on?
Interoperability is very unlikely to include sharing common settings in the way you desire. It's more likely to be about sharing data between applications.
As for Firefox/KWallet integration: I admit it would be nice for some (personally I use my head to remember passwords - it's more portable). It's not impossible but it may be impractical.
KWallet has a published API and applications that integrate with it adhere to that API. I've not seen many non-KDE applications that do so. Mozilla could probably integrate with it but would they want to stop at KDE support for this? Firefox is deliberately platform-independent ( granted it does use the host-system's dialogs ). Besides, I'm not so certain than the other platforms it can run on have an equivalent to KWallet. I don't recall Windows, OS X or GNOME having a central app which permits password storage across _all_ (inclusing non-native) applications.
cheers
Ryan
crimperman
16 years 10 weeks 1 day 41 min ago
FUD
I think this is pretty much FUD. Granted it's not from Microsoft or another proprietary company but it does take the approach of "we can't money from this unless we change the model" - which has never really been the point of free software in particular.
FTA: "Microsoft has thousands of developers working in lockstep on getting all of its applications to work together. There's never a worry that Microsoft Office will function effortlessly with Microsoft Exchange, Microsoft SharePoint, Microsoft Dynamics and so forth. Can the same be said across open-source applications? I would say the answer to that question is: not yet."
This is basically his premise: "Microsoft's products work together perfectly-so should OpenSource ones".
I have two issues with this:
1. MSOffice and Exchange do not always function effortlessly together. Take a look at Experts Exchange or similar and see how many issues there are around interoperability betwene MS products.
2. Open source products are not made by the same company so to expect the same standard (even though MS doesn't achieve it) is unfair. Do MS SQLServer and Oracle integrate seamlessly? Do MS Office and say Adobe Acrobat (not reader)? Then why should we demand that Open Source applications do?
Aside from that adherence to open standards by all would pretty much guarantee interoperability - certainly on a data front.