Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.os.vms    |    DEC's VAX* line of computers & VMS.    |    264,096 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 263,544 of 264,096    |
|    =?UTF-8?Q?Arne_Vajh=C3=B8j?= to John Dallman    |
|    Re: VMS previous DEC/CPQ/HP[E] decisions    |
|    13 Oct 25 16:52:32    |
      From: arne@vajhoej.dk              On 10/11/2025 10:13 AM, John Dallman wrote:       > In article <10cdflq$5c0$2@reader2.panix.com>,       > cross@spitfire.i.gajendra.net (Dan Cross) wrote:       >> I think also the focus shifted dramatically once Java came onto       >> the scene; Sun seemed to move away from its traditional computer       >> business in order to focus more full on java and its ecosystem.       >       > They tried that on us, but were deeply unconvincing.       >       > They were expecting us to be impressed that they'd done JNI wrappers of       > about ten functions from our 500+ function API. We said "Presumably you       > have tools to generate this stuff automatically?" and they didn't       > understand what we were talking about.              You may have been lucky.              :-)              The concept of:              Java code---(millions of low level function calls via JNI)--->native code              is not good.              Java code---(thousands of high level service calls via JNI)--->native code              may work OK.              Moving data between managed code and unmanaged code is in general       tricky and cost a lot of CPU cycles.              And Java JNI is not even a good implementation of that.              .NET did much better with InteropServices/DllImport and C++ CLI.              Arne              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca