Hacker News new | past | comments | ask | show | jobs | submit login

Oracle's newer VM patents.



Cleverly forgetting those from IBM, Intel, Microsoft, Amazon, Azul, SAP, Alibaba and everyone else that works on Java?


No, because the core patents behind the features you're asking about are very much in Oracle's possession, as those class library improvements were overwhelmingly written by Sun/Oracle engineers. Those other actors you're talking about absolutely contribute, but not in as nearly a fundamental to the implementation of the library and runtime definitions kind of way.


Are you now asserting that Sun/Oracle has pattents regarding Azul's GC implementation, PTC real time threading and AOT compiler, Aicas hard real time GC, IBM mainframe implementations of their own JVM and WebSphere Real Time VM?


No, I'm asserting that those (while being extremely valuable additions to the field) have nothing to do with a newer version of Java which is what you were asking about.

> the next excuse for not updating Android Java to latest versions.


So let's back to that.

The only two companies that ever had issues with Java owners were Microsoft thanks to J++ and Google with its Android Java.

Lots of companies have produced their own Java version, without any issue.

Sad that things have gone this way and Google isn't being hardly punished for what they did to Java ecosystem with their special flavoured Java and by giving Sun's fatal blow.

Thankfully they never thought of buying Sun to own Java, as it would have been yet another project on their graveyard, looking to how they handle it on Android.

But yeah lets not get distracted and join the Oracle hate mob.


> The only two companies that ever had issues with Java owners were Microsoft thanks to J++ and Google with its Android Java.

While the Apache project is not a company, it definitely had issues with Sun over Apache Harmony's Java certification (I'm not confirming or denying that this was part of IBM's proxy wars with Sun).

Incidentally, Apache Harmony is what Google used to jumpstart not-officially-Java compatibility with Android.

> Sad that things have gone this way and Google isn't being hardly punished for what they did to Java ecosystem with their special flavoured Java and by giving Sun's fatal blow.

Speaking of "special flavoured Java" - what Google did with Android was an improvement over the fragmented mess that was Java ME. Good God, I remember Sun-approved, vendor-specific extension, so Samsung/Nokia/Sony-Ericsson each had its own API for the same functionality - e.g. checking for connectivity. Additionally, the API could be subtly different for each phone model from the same vendor. You either had to create different build pipelines (1 per vendor, with overlays/facades), or do an ungodly amount of reflection in your runtime logic. So much for "Write Once". Sun was never interested in licensing Java SE on mobile, I don't see how Google could have killed them when they weren't even competing on "your phone as a full-blown computer"

I loved Sun Microsystems; they were idealistic, geeky, and made very cool, albeit expensive gear. I suspect their idealism made them hop onto the open source bandwagon without due consideration for long-term sustainability. They open-sourced the OS (OpenSolaris), open sourced the application stack (Java, perhaps reluctantly), while their server hardware was being disrupted by x86; how were they going to make money? They were not long of this world - Linux, commodity x86 and open source Java Application Servers killed Sun. IBM was in a similar place at the time, and they shifted focus to consulting.

> But yeah lets not get distracted and join the Oracle hate mob.

I've been part of that mob for a long time, fuck Oracle for how they treated Gosling, how they handled the JCP, and mishandled OpenOffice (until it blew up in their face). I didn't care for MySQL, but it seems to have been ok so far.


I love how Google apologists sell Android fragmentation and OEM modifications to AOSP as an improvement over J2ME.

Now excuse me while I will compile some stuff in Java 16 with the SIMD preview API.


If you had the displeasure of coding a moderately complex J2ME app - you'd know this it's not mere apologia. While Android fragmentation is terrible (and was worse in the android 1 era), J2ME was at least 10x worse - I haven't met a single person who disagrees with the idea that Android was a massive improvement over J2ME - I noticed not even you are making that argument, you're only chiding me for bringing it up.

Sun could have fixed it by availing Java SE on mobile devices and forcing vendors to implement the same APIs in the same java.smartphone.* packages - (they already had a Java TCK!), but they were not prepared for the smartphone era, sadly. Look at the CLDC[1] classes and tell me how one could write apps that could possibly compete with the iPhone apps in 2007.

1. Sadly - but predictably, Oracle nuked java.net which had documentation on CLDC 2.x/MIDP, I managed to find CLDC 1.1 documentation, the classes are surprisingly sparse, so J2ME was probably worse than I remember. https://docs.oracle.com/javame/config/cldc/ref-impl/cldc1.1/...




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: