Home

News: Oracle discontinues free Java time zone updates

  1. For a long time in the Java world, there has been a free tool called the "tzupdater" or Time Zone Updater released as a free download first by Sun and then Oracle.  This tool can be used to apply a patch to the Java runtime so that time zone information is correct.  This is necessary since some time zones in the world are not static and change more frequently than one might think; in general time zone updates can be released maybe 4-6 times a year.  The source information backing the Java timezone API comes from the open source Olson timezone database that is also used by many operating systems.  For certain types of applications, you can understand that these updates are mission critical.  For example, my company operates in the private aviation sector so we need to be able to display the correct local time at airports around the world.


    So, the interesting part is that Oracle has now decided to only release these updates if you have a Java SE support contract.  See the following link:
    http://www.oracle.com/technetwork/java/javase/downloads/tzupdater-download-513681.html
    Being Oracle, such licenses are far from cheap.  

    In my opinion, this is a pretty serious change in stance for Oracle and amounts to killing free Java for certain types of applications, at least if you care about accuracy.  We are talking about the core API class java.util.TimeZone.  This begs the question, can you call an API free if you have to pay for it to return accurate information?  What is the point of such an API?  Should the community not expect that core Java classes are fully functional and accurate?  I believe it is also a pretty bad move for Java adoption for these types of applications.  If my company as a startup 10 years ago would have been presented with such a license fee, we almost certainly could not have chosen Java as our platform as we could not afford it.

    Threaded Messages (21)

  2. Oracle and Java[ Go to top ]

    Hi Noel -<br>

    (I also responded to your email.)<br>

    I asked about this internally. This is the response I received back:<br>

    <blockquote><i>We update the timezone data regularly every time we ship a new Java update (6 times per year), and the most recent version of Java is always royalty-free.<br>
    We consider the TZupdater to be a support tool for our long-term support customers and we do charge for support as a way to fund the development of Java. We don't charge for it specifically, it's just part of our overall support offering.</i></blockquote><br>

    Obviously, the charging for support brings up a complex topic, in that Oracle wants to encourage the continued growth and usage of Java, yet also needs to find ways to pay for the maintenance work on Java -- which is quite expensive, as you could imagine.<br>

    Peace,<br>

    Cameron Purdy | Oracle<br>

    <i>For sake of full disclosure, I work at Oracle, including managing the group responsible for the development of the Java EE platform. The opinions and views expressed in this post are my own, and do not necessarily reflect the opinions or views of my employer.</i>

  3. TSS and formatting[ Go to top ]

    Wow! One week to the next here you have to learn completely different ways to get your message to format correctly!

  4. TSS and formatting[ Go to top ]

    If you want to be able to receive the latest information on TSS formatting rules, please buy support from TSS. Money you pay will be used to keep this site up-to-date :)

    Regards,

    Ivan

  5. Oracle and Java[ Go to top ]

    I did see your email reply today Cameron.  I appreciate you taking the time to look at the issue and reply.

  6. Oracle and Java[ Go to top ]

    FWIW, I updated the information at the link from the origional article.  Admittedly, the origional information was sparse, so I hope the expanded info helps.  Also, FWIW, I sense a misunderstanding in the origional post -- we always provide the most current TZ information in every single update.  This tool is not required to get TZ updates, it's always updated with every Java update - gratis.

    Disclaimer - I'm in the Java SE PM team.

  7. Oracle and Java[ Go to top ]

    Hi Donald,

    I applaud the fact that Oracle ships every Java update with the latest time zone information, but there are two big problems with doing just that and this is why I assume the tzupdater tool has always existed.

    1. Updates to Java are not released every time an Olson time zone update occurs.  So, the two release cycles are not in sync.  Imagine a user of your application calling in to report that the local time that the application is displaying is wrong.  Are you going to tell them to just wait for the next release of Java SE ?
    2. For most companies / open source projects, updating the version of Java that your application runs on is a fairly major initiative.  It involves a lot of regression testing and quality control across numerous environments.  You should not have to go through this process to ensure that Java can display the correct local time.
  8. Oracle and Java[ Go to top ]

    Sorry, but updating the jvm should be an easy task in any project. Can you cite any specific story where a minor jvm update introduced a regression in your project? "A lot of regression testing" is just an exageration.

  9. Oracle and Java[ Go to top ]

    Sorry, but updating the jvm should be an easy task in any project.

    Obviously you've never worked with large insurance or bank companies. They won't let you use any version, only the ones that passed their quality tests. These tests costs them money and they won't upgrade unless they have a very good reason to. I know banks that still use 1.5 jdks (not the latest one) and the sdk revision hasn't changed in the last 4 years.

    Can you cite any specific story where a minor jvm update introduced a regression in your project? "A lot of regression testing" is just an exageration.

    On the top of my head: the infamous 6u21 when Oracle changed the company value from Sun to Oracle, which broke Eclipse support. I had some breaking regressions in Java Web Start also (don't remember on which release). When you have servers which run dozens of applications, a lot of regression testing is in fact quite an understatement.

  10. Oracle and Java[ Go to top ]

    Sorry, but updating the jvm should be an easy task in any project.

    Obviously you've never worked with large insurance or bank companies.

    You're just wrong, that's exactly what I do.

    They won't let you use any version, only the ones that passed their quality tests. These tests costs them money and they won't upgrade unless they have a very good reason to. I know banks that still use 1.5 jdks (not the latest one) and the sdk revision hasn't changed in the last 4 years.

    Those are things I sometimes have to fight. I always won this: upgrading is essentialy trading known jvm issues for risk. A minimum of testing is required, but 1) not that much, and 2) you should have automated tests anyway which at least validate most features.

    Upgrading should be planned, but is something to do on a regular basis.

    Can you cite any specific story where a minor jvm update introduced a regression in your project? "A lot of regression testing" is just an exageration.

    On the top of my head: the infamous 6u21 when Oracle changed the company value from Sun to Oracle, which broke Eclipse support. I had some breaking regressions in Java Web Start also (don't remember on which release). When you have servers which run dozens of applications, a lot of regression testing is in fact quite an understatement.

    Sorry, but this didn't break production. Eclipse is just a broken tool, you'll have such issues all the times. At least, that's "just" your development environment. Such a problem is detected long before it can do any real damage.

  11. Oracle and Java[ Go to top ]

    Hi Yannick,

    It is not coming but minor version upgrades to the JVM can cause issues, just as another poster mentioned.  The last time we got bit by this was the following:

    http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7105007

    As the other poster also mentions, many companies spend a lot of time and effort around verifying JVM version upgrades.  Irregardless if the effort is in testing or just getting the right JVM version on every developer, test, staging, production, etc. machine, it is more effort than should be necessary for Java to produce a correct local time.

  12. Oracle and Java[ Go to top ]

    Oracle wants to encourage the continued growth and usage of Java, yet also needs to find ways to pay for the maintenance work on Java -- which is quite expensive, as you could imagine.


    Bleah. Oracle wants to make as much money as possible, period. It's in their DNA.

  13. Oracle and Java[ Go to top ]

    Obviously, the charging for support brings up a complex topic, in that Oracle wants to encourage the continued growth and usage of Java, yet also needs to find ways to pay for the maintenance work on Java -- which is quite expensive, as you could imagine.<br>

    Peace,<br>

    Cameron Purdy | Oracle<br>

    Buyer's remorse? So Oracle decides to arbitrarily add a vig after the fact. Sort of bait and switch methinks. 

    Another nail in the Java coffin. So sad with all the new stuff in the Java pipeline. CIOs and CTOs are already sour on their java installations, they're gonna luv this.

    This has a wider impact than even the reasonable arguments in this thread are bringing up. What about B2B issues? Small client A decides to go GF/openJDK while Big Boy B has the enterprise license setup with deep pockets for support. There is an API between the two that is dependent on reliable time calculations based on global location and timezones -- heck, it doesn't even have to be an API, just end of day file for clearing with time calculations in the stream. Now client A is forced into a support agreement because they need to sync with B2B partners when originally the package was presented as a free download? As Noel mentions, if this was relayed honestly up front, perhaps other operational decisions would have been adopted.

    This is a core functionality issue. I can see this(sort of) with JMS,JDBC and so forth. Not basic time calculations. What next, everyone will be forced into additional charged support to handle commas or backslashes in the java.lang.String API?

    Why not just have a rolling Kickstart program if a multi-billion dollar company can't afford to maintain their IP portfoliio. 

    Does this include entities who already purchased other products such as Weblogic or the Oracle Edition of Glassffish? Or will this be add-on to those installations too? 

    As a final question, when is the timeline for this change. I guess this will have to be relayed to clients so they can start plans to move off the java platform. 

  14. Oracle and Java[ Go to top ]

    Frank -

    This is a core functionality issue. I can see this(sort of) with JMS,JDBC and so forth. Not basic time calculations. What next, everyone will be forced into additional charged support to handle commas or backslashes in the java.lang.String API?

    I think that is a reasonable example. Here is my understanding: If there were a bug in java.lang.String, Oracle would release a new (also free) version of Java with a fix to that bug. Similarly, when time zone changes occur, Oracle releases a new (also free) version of Java with those time zone changes incorporated. If someone wants a back-port of one of those fixes or changes to apply to an earlier version of Java (as is the case with the tzupdater utility), they would need to have a support contract.

    Why not just have a rolling Kickstart program if a multi-billion dollar company can't afford to maintain their IP portfoliio. 

    This is an odd question. I am not sure how to respond in a logical manner. Are you focusing that comment on Oracle, or on all of the companies (e.g. Redhat with Linux & JBoss, VMWare with Spring, etc.) who charge for support.

    Regardless, I see no reason why you or someone else could not provide -- for free -- a tzupdater utility of your own for each change that occurs to time zone information, although it would be a little difficult to ensure that it is tested with and supported on the various thousands of hardware, software and Java version combinations that Oracle tests on and supports.

    Does this include entities who already purchased other products such as Weblogic or the Oracle Edition of Glassffish? Or will this be add-on to those installations too? 

    To the best of my knowledge, those products include Java support for the use of those products.

    As a final question, when is the timeline for this change. I guess this will have to be relayed to clients so they can start plans to move off the java platform. 

    From the original post, it sounds like this change has already been rolled out.

    Peace,

    Cameron Purdy | Oracle

    For sake of full disclosure, I work at Oracle. The opinions and views expressed in this post are my own, and do not necessarily reflect the opinions or views of my employer.

  15. Oracle and Java[ Go to top ]

    Frank -

    This is a core functionality issue. I can see this(sort of) with JMS,JDBC and so forth. Not basic time calculations. What next, everyone will be forced into additional charged support to handle commas or backslashes in the java.lang.String API?

    I think that is a reasonable example. Here is my understanding: If there were a bug in java.lang.String, Oracle would release a new (also free) version of Java with a fix to that bug. Similarly, when time zone changes occur, Oracle releases a new (also free) version of Java with those time zone changes incorporated. If someone wants a back-port of one of those fixes or changes to apply to an earlier version of Java (as is the case with the tzupdater utility), they would need to have a support contract.

    Why not just have a rolling Kickstart program if a multi-billion dollar company can't afford to maintain their IP portfoliio. 

    This is an odd question. I am not sure how to respond in a logical manner. Are you focusing that comment on Oracle, or on all of the companies (e.g. Redhat with Linux & JBoss, VMWare with Spring, etc.) who charge for support.

    Fair enough but your posted email seemed like a hat in hand rationale to charge for tzupdater. So are you saying that these other corporations are distributing "incorrect" free releases of their products and telling their customers that they need to pay for the "right" versions. The argument on this thread is sort of like Redhat saying, "it's too expensive to maintain Linux you know,so in order to maintain the community because it's expensive you know, ntpdate has to be paid for." As far as distributing "free versions" with the updates, as evidenced by the recent security disasters, the update cycle may be a liability to some shops -- and in fact makes paid support's value up for debate. But that's a whole other discussion.

    Regardless, I see no reason why you or someone else could not provide -- for free -- a tzupdater utility of your own for each change that occurs to time zone information, although it would be a little difficult to ensure that it is tested with and supported on the various thousands of hardware, software and Java version combinations that Oracle tests on and supports.

    I thought the lib/zi tree and ZoneInfo format was Oracle property and would violate EULA's and copyrights if modified or used?. Can you confirm this isn't true?

    Does this include entities who already purchased other products such as Weblogic or the Oracle Edition of Glassffish? Or will this be add-on to those installations too? 

    To the best of my knowledge, those products include Java support for the use of those products.

    Could you relay this sentiment to your AutoVue group? They distribute 1.6.0_31 with the security holes in all their glory. Asking for updates (just to the JRE) require secondary support contracts. If we use a self-installed JRE latest version, it violates the AutoVue proper support contract -- a catch 22 situation. (NB: There are both OEM and End User Support Contracts paid for so there is no "free lunch" request here). It's this inconsistency across the middleware products which brings the frustration.

    As a final question, when is the timeline for this change. I guess this will have to be relayed to clients so they can start plans to move off the java platform. 

    From the original post, it sounds like this change has already been rolled out.

    Good to hear. Let's see what comes next. In Oracle's defense, maybe the time has come that Java should not be open or free anymore if even Oracle has financial trouble supporting it. It's a little puzzling because your CFO has been stating phenomenal earnings/margins across your entire IP portfolio so this revelation that maintaining Java is becoming expensive should bring up some interesting questions on the earnings conference call later this month. 

    Peace,

    Cameron Purdy | Oracle

    For sake of full disclosure, I work at Oracle. The opinions and views expressed in this post are my own, and do not necessarily reflect the opinions or views of my employer.

  16. Oracle and Java[ Go to top ]

    Hi Frank -

    Fair enough but your posted email seemed like a hat in hand rationale to charge for tzupdater. So are you saying that these other corporations are distributing "incorrect" free releases of their products and telling their customers that they need to pay for the "right" versions. The argument on this thread is sort of like Redhat saying, "it's too expensive to maintain Linux you know,so in order to maintain the community because it's expensive you know, ntpdate has to be paid for." As far as distributing "free versions" with the updates, as evidenced by the recent security disasters, the update cycle may be a liability to some shops -- and in fact makes paid support's value up for debate. But that's a whole other discussion.

    I think you may have misunderstood what I was saying. The latest Oracle JDKs and JREs (which are provided for free) are supposed to always ship with the latest timezone info included. That means that when there is a change to timezone information, it would be possible to get that latest timezone information by downloading the latest JDK or JRE (once that information has been incorporated).

    It sounds like you thought Oracle was purposefully shipping incorrect timezone information, and then charging to fix it. This is not correct, although it is possible that the latest timezone information wasn't always included with the latest JDK and JRE downloads. (The Java team is currently reviewing their processes to ensure that this type of problem will not occur in the future.)

    Regarding the recent security fixes, I think Oracle has demonstrated quite a commitment to fixing these, and has continued to provide significant updates (for free) even to product versions that are beyond their stated "end of life" dates. This was not a small effort -- literally hundreds of people were involved for over a year now in attacking the entire set of known security issues, and also working to evaluate security over the entire range of Java libraries. Fortunately, the team is finally finishing the bulk of this project.

    Peace,

    Cameron Purdy | Oracle

    For sake of full disclosure, I work at Oracle. The opinions and views expressed in this post are my own, and do not necessarily reflect the opinions or views of my employer.

  17. Trying to make java paid[ Go to top ]

    Here another move from Oracle trying to make companies to paid for something cheap an expensive price. Not even Microsoft does this kind of things with their platform.

    Oracle may have a different opinion, but I see this move from this perspective:

    Oracle cannot change java license from free to paid, the whole community will burn the offices of Oracle with Larry Ellison inside. So what is left? Well you can change things that doesn't impact the developer community but companies, and companies are used to pay for support. Then we take it data and wrap it as support and we make sure nobody can access the data when needed unless you paid for it.

    This is one trial to make companies pay for using Java, as I think most of the developers in the community, I thinks this is a step in the wrong direction. In my opinion Oracle should burn money and effort on creating a java application market, where developers and companies can sell online java applications, APIS, components, etc.

    Best regards

  18. fett verbrennungs ofen[ Go to top ]

    fett verbrennungs ofen

  19. FYI -- As per Henrik's blog [1], the TZUpdater tool will be made available, likely by EOB pacific today.  Much thanks to all who provided constructive feedback and apologies for any inconvenience.

     - Don

    [1] - https://blogs.oracle.com/henrik/entry/tzupdater_for_jdk_7_available
    Disclaimer - I work for Oracle, and am on the Java SE PM team.

  20. FYI -- As per Henrik's blog [1], the TZUpdater tool will be made available, likely by EOB pacific today.  Much thanks to all who provided constructive feedback and apologies for any inconvenience.

     - Don

    [1] - https://blogs.oracle.com/henrik/entry/tzupdater_for_jdk_7_available
    Disclaimer - I work for Oracle, and am on the Java SE PM team.

    Well, that makes it much simpler!

    Peace,

    Cameron Purdy | Oracle

    For sake of full disclosure, I work at Oracle. The opinions and views expressed in this post are my own, and do not necessarily reflect the opinions or views of my employer.

  21. Hi Donald,


    Thank-you and Cameron for taking some time on this and taking this seriously.  I appreciate that Oracle is doing the right thing here.

  22. Wise decision[ Go to top ]

    We celebrate the news!

    Thanks a lot