Archive for September, 2009
What began as a story of evil Google seeking a monopoly on Android apps has become a kerfluffle.
The opening shot was a cease-and-desist letter issued by Google against Steve Kondik, aka Cyanogen, Kondik was producing a modified Android ROM that included proprietary Google applications.
While Android is open source, it does support proprietary extensions. Google has provided its services, like GMail, to Android in this form.
Reaction was immediate. Google wallops Android developer was typical. Has Google shot itself in the foot was a typical question.
Indeed I was all ready to write a screed against trying to turn open source proprietary before I did the research.
First, Google has a good reason for wanting to retain control over its own services. This lets it retain legal cause against malware aimed at its servers using its software.
Second, after some hemming-and-hawing, Kondik backed down. He said he would provide a work-around so his Cyanogenmod will still work with Google applications. A group of college kids called the Open Android Alliance offered to do a purely open source Android.
There was even some action inside Google itself, with one developer considering resigning but the thinking better of it. “Let’s try to move on and be constructive and talk about how we can make Android better for everyone,” he concluded.
So why is this a kerfluffle? Why isn’t it a scandal?
Because so far only one lawyer has made a dime off this, the Google lawyer who wrote the original letter. (And he or she is on salary — if they hadn’t written this they might have just played ping-pong on the Google dime.)
In the end this was an open source dispute, handled by open source developers in an open source way. Let’s find a workaround. Let’s find a way to get along. There ain’t no one here but us chickens.
Filed under: Office, Web services, Google, Web
If you’re not a developer, you’re probably not aware that Google Docs has an API available for various document-related services. Recently Google added a new feature that allows developers to create applications that will pass an image-based (.png, .jpg, or .gif) document to the API, and using OCR (Optical Character Recognition) technology, generate and pass back an editable text-based document. Currently the service can handle documents up to 10 MB in size.
For non-developers, this is certainly interesting, but not terribly useful. Well, a live demo is available that will allow you to test the service yourself. At the time of this writing I was successfully able to convert a single page document, though there are reports that the service has become overloaded and is slow or sometimes fails to respond. Given that this is a demo, that’s probably not terribly surprising.
If Google is exposing this functionality to developers, it seems at least possible that it could become a built-in feature of Google Docs for regular users at some point in the near future. With Google’s recent acquisition of reCaptcha, it seems likely that Google’s document-scanning capabilities will soon be better than anyone else’s. That would make for an incredibly powerful feature for Google to offer its users.
[via Google Operating System]
Google Docs API now provides OCR service originally appeared on Download Squad on Wed, 30 Sep 2009 08:00:00 EST. Please see our terms for use of feeds.
Read | Permalink | Email this | Comments
Sponsored Topics:
Google Doc – Optical Character Recognition – Google – Download Squad – Search Engines
Go to Source
Is there any question map mashups have gone serious? How about billions of dollars serious? The U.S. website Recovery.gov was built to make transparent the use of economic stimulus money. Now you can track every dollar on a map.
Click on Where’s The Money Going and you’ll get a Flash map with all sorts of options for slicing and dicing the data. See all funding by state, department, or broken down by specific agency. The awards types are also shown separately, so you can tell a contract from a grant or loan.
Sunlight Labs, which has a mission of “opening America’s government,” is not particularly impressed yet:
“The verdict? Well, it’s hard to say — the site’s a bit broken. There are 404s all over the place, most gallingly on the data download page.”
Sunlight’s Tom Lee also explains that much of the data is aggregated from other government sources previously available. Not that it’s a valid excuse for not allowing the download. You can get at some of the data. For example, within the map view, a KML file is provided, which is handy.
The map is built on top of ESRI technology. The company’s GIS software is used throughout all levels of government, so it’s not surprising to see it on a .gov site. Though it is courting the mainstream mashup developers, ESRI’s biggest advantage versus Google Maps and other mapping providers is that ESRI is seen as the platform to use if you’re serious.
Hat tip: ComputerWorld
Related ProgrammableWeb Resources
So, here we are: we have just completed our third Google Summer of Code™. Despite the nostalgia that has settled in after the end of the summer, we are all feeling very, very happy about how things went this year. While observing my fellow mentors, who are busy integrating our students’ contributions into our code base, I am tempted to reminisce about our three year history with the program and the lessons we’ve learned.
First of all however, a quick history of our participation: Our adventures started in 2007 when we were accepted into the program for the first time. I can still remember jumping all around the room when I saw SIP Communicator’s name in the list of accepted organizations. Back then we were a brand new project and this acceptance was a tremendous recognition. As it turned out later, it made a great difference in terms of popularity, credibility, and bringing new contributors both directly and, above all, indirectly.
Our first summer went exceptionally well. First of all, we had a decent number of applications, 87 to be precise, and by decent I mean not too much for the available mentors to handle and yet enough for us to have a wide choice of candidates. We received funding for eight student projects, which, as it turned out later, was also just right. At the end of the summer we had 7 successful students. During the months following Google Summer of Code 2007, we integrated virtually all the work that came out of it. We also voted and accepted two of the students as permanent committers.
Then came 2008. Once again we were all rejoicing in anticipation of a productive summer. This time we had 187 candidates and were received funding for 20 student projects. At that point, however, we started realizing how big a number 20 is and we got a bit scared. We were afraid that would be too many students for us to handle so we decided to only take 15 and let other projects mentor the additional five. It turned out later that 15 was still a bit too many – more on that later. The summer went pretty well and a lot of work got done. Once again we voted and accepted two of the students as permanent contributors and only had a single failing student at the end of the season.
This leads us to 2009, and boy was this year a good year! I can now safely say that this has probably been our best participation so far. We received a staggering number of applications: 203 to be precise. We only had around 15 mentors so it took us quite some time to go through all of them. Once we were done with the evaluations we requested 12 student projects but played it safe and decided to go with only ten, leaving the rest of the funding for other student projects. Once again, it was a really great summer. We are still in the process of integrating all the contributions and it will probably take us a few months before we are done. Even at this point, with about 30% of the work in our repository, we have already voted and accepted 2 of the students as permanent committers with probably two or three more to come in the the following months. Hip Hip … Hoorray!!!
Here’s an in-depth look at some of our 2009 projects:
Growl Notifications, and Next Generation Sparkle Updates
Egidijus Jankauska from the United Kingdom has implemented a native popup notification for the MacOS X version of SIP Communicator. It makes use of the Growl notification daemon through a new implementation of the Java bindings of the Growl API. For that purpose, Egidijus has implemented a dynamic library using Java Native Interfaces, a set of Java interfaces, and the corresponding implementations for SIP Communicator. The new born library can of course be used in other projects and this implementation has already been integrated in our source trunk.
Egidijus has also updated our package update system on MacOS X. It was based on Sparkle 1.1, and Egidijus has provided the necessary patches and documentation to switch to Sparkle 1.5b6. This work has also been integrated in our source trunk.
Egidijus has since been voted as a committer and is now part of our developer team!
Hush-hush Chats with Off The Record (OTR) Messaging
George Politis from Greece worked on extending SIP Communicator with Off The Record (OTR) message encryption. OTR provides encryption, authentication, deniability, and strong forward secrecy. Until now SIP Communicator did not have any text message encryption and our chats were often unprotected. George started with the implementation of our own Open Source native java OTR library, which can also be used in other projects. George also implemented all the message transformation functionalities and the GUI necessary for us to integrate OTR support in SIP Communicator. It is already implemented in many of the other popular instant messengers such as Kopete, Pidgin, Adium, mICQ, Miranda, and Trillian. SIP Communicator is now able to carry out encrypted communications with other SIP Communicator clients and the aforementioned messengers.
George’s implementation has already been integrated in our source trunk and George has achieved committer status for SIP Communicator with a strong approval of our community.
Storing Chat History and Contact Lists in a Database
Ajay Chhatwal from India was in charge of implementing a Database system to allow us to store all chats in a database instead of XML files. Ajay has studied many database systems, produced a comprehensive comparative evaluation on them and suggested a winner that would best suit our use case. He has then implemented a database service and a backend to provide a working database service to all the components of SIP Communicator, after which he worked on a transition mechanism that would allow transferring XML files from the old implementation into the new database system.
Once he completed his work on the history modules – yes, he still had time to hack before the end of the summer – Ajay has also coded a new version of the contact list service which now also uses the database service.
We’re hoping to vote Ajay in as a committer soon.
Recognizing and displaying remote user agents
This one was worked on by Brett Geren from the United States. The project consisted of retrieving the names of the applications that our buddies are using when chatting with us, and showing the application icons to the user. In order to accomplish this task, Brett first completed extensive research determining which of the protocols we support in SIP Communicator actually deliver such information and how they transport it.
He then defined the interfaces necessary for a new user agent module and implemented the feature for MSN, IRC and XMPP. During the second half of the program, he worked on the user interface that actually displays the remote client icon and allows users to configure the behaviour of the user-agent plugin. He also completed tests with a long list of known clients in order to confirm the way they are publishing their client name and to make sure that SIP Communicator was working with them as expected.
Ed. Note: This post is the first installment from the SIP Communicator project on their participation in the Google Summer of Code program. Look forward to even more information on their 2009 student projects and some in-depth details on lessons learned on this blog next week. Stay tuned!
By Emil Ivov, Project Lead, SIP Communicator and Google Summer of Code Mentor
If everything were open source, or one open source implementation controlled a market, would that be an lllegal monopoly?
(Picture from the Tiger Woods Foundation.)
Antitrust law is about more than market share. It’s about using that market share to destroy competition. A monopolist must take an overt action against another company before the Justice Department is going to take a look at it.
IBM’s mainframe monopoly in the 1950s was followed by IBM using its control of designs to keep competitors out of its market. AT&T’s telephone monopoly in the 1960s was followed by its trying to prevent the creation of MCI. Microsoft’s monopoly in the 1990s was followed by its bundling Internet Explorer to destroy Netscape.
You have to control the market and use that control to abuse other entrants before lawyers toss the monopoly grenade at you.
Let’s put it this way. The Bill and Melinda Gates Foundation is large and powerful. It is bigger, relative to its foundation brethren, than any other charitable foundation ever.
Now Tiger Woods also has a foundation. It had an event near my home yesterday. It involved teens, and goal-setting. The only way for Tiger to have a beef against the Gates folks would be if they tied-up all the teens in the country, actually preventing Tiger from helping any.
I doubt Tiger would be upset in that case. It’s only if he’s prevented from helping anyone by the dominant foundation’s actions that he might have call to become angry. It’s not going to happen.
An open source monopoly is simply not a barrier to competition. Even if a GPL program had market dominance you could still compete with it.
The whole question is silly except for our friend Matt Asay’s realization that foundations are a stable form of open source governance. I have no monopoly on that. I don’t even have a business model for it. All I have is what Tiger would offer if the Gates Foundation solved all teenagers’ problems.
An attaboy.
Filed under: Productivity, Web services, Google, Education
Google must have been reading our back to school series, because the Google Docs interns have added a bunch of new features designed specifically for students. There’s something for everyone, whether you’re writing papers or doing problem sets. Although Google is highlighting thiese improvements as useful to students, some of them could definiteltly come in handy in a nonacademic situation.
For those of you taking math classes, there’s an equation editor, so you can actually take math notes on your laptop without a special app. Chemistry students will be happy to note the addition of superscripts and subscripts, making it easier to put the 2 in H2O.
Google Translate is now available from within Docs, which is of general interest, but should appeal specifically to language students. For students conducting surveys, there’s now an option to proceed to a different question depending on a participant’s answer. If you’re just writing a paper, you can take advantage of additional bullet styles and the ability to export footnotes as endnotes if necessary.
Google Docs goes back to school with new features originally appeared on Download Squad on Tue, 29 Sep 2009 08:00:00 EST. Please see our terms for use of feeds.
Read | Permalink | Email this | Comments
Sponsored Topics:
Google Doc – Download Squad – Google – Web service – Searching
Go to Source
Fast on the heels of Sun’s release of an updated version of the MySQL Enterprise Monitor, MySQL has released the results of a survey of 637 small and medium-size businesses (less than 500 employees) in Europe documenting open source usage.
Several enterprise mashup proponents, including ProgrammableWeb, have come together to form the new Open Mashup Alliance (OMA). The OMA has been founded with the goal of supporting the implementation of enterprise mashups along with an open language that promotes enterprise mashup interoperability and portability.
The OMA’s founding members companies include: Adobe, Bank of America, Capgemini, Hinchcliffe & Co., HP, Intel, JackBe, Kapow Technologies, Programmable Web, Synteractive, and Xignite.
This is very promising development in the enterprise mashup space, a market we have regularly covered here on ProgrammableWeb. As you can see in the list of founding companies, many of them are companies that we have identified as leading the charge in the implementation of enterprise mashups.
In order to put some substance behind the goals of the OMA, one of the major proponents of enterprise mashups, JackBe, have contributed its Enterprise Mashup Markup Language (EMML) and runtime engine to the OMA as an open source resource (licensed under Creative Commons). The idea behind EMML is that it will break down the barriers for implementation of enterprise mashups by providing a common language for development of the mashups and a corresponding engine that processes mashup scripts written in EMML. You can see a snippet below:
You can learn more about the OMA on its FAQs page, which also outlines why organizations may be interested in joining the OMA (membership is open to anyone). OMA has made documentation for EMML available online, and you can download the EMML schema and runtime engine as well.
Michael Ogrinz, principal architect at Bank of America and an occasional contributor to ProgrammableWeb, perhaps sums it up best in the OMA’s introductory press release:
“For enterprise mashups to take hold, we need to remove the ‘vendor lock-in’ concerns raised by today’s proprietary toolsets. We also need to inspire the innovative minds of the open-source community to start working in this space. By establishing an open standard for mashups, the OMA and EMML addresses both of these issues.”
You can also read Mike’s blog for some of his additional insights on the OMA.
We’re looking forward to contributing to the OMA’s efforts and encourage others to join the OMA in working towards an even better programmable web.
Filed under: Audio, Utilities, Video
Some YouTube videos are really only worth it for the audio portion. I’ve seen tons of songs posted to YouTube with still images instead of real videos, and often wished I could just download the audio in one step. Dirpy lets you easily save the sound from a YouTube video as an mp3 for later listening, or save the whole video if you really want to.
One of Dirpy’s best features is the ability to specify the portion of the audio you want to download by entering start and stop times, which really helps when you come across a video with a long intro. Dirpy also lets you set ID3 tags for your ripped audio files, so they’re probably labeled in iTunes and other players. You can find videos to rip by searching on Dirpy or using the bookmarklet on any YouTube video page.
[via Lifehacker]
Dirpy rips YouTube audio to mp3 originally appeared on Download Squad on Mon, 28 Sep 2009 08:00:00 EST. Please see our terms for use of feeds.
Read | Permalink | Email this | Comments
Sponsored Topics:
Download Squad – YouTube – MP3 – ITunes – Uploading and downloading
Go to Source
Data from the Coverity Scan Open Source Report found a 16 percent improvement in the quality of open source projects actively participating in the scan. This is exactly the type of data that open source vendors and proponents want in their back pockets. But is it accurate? Hold your tomatoes and let me explain.