Understanding through Discussion


Welcome! You are not logged in. [ Login ]
EvC Forum active members: 77 (8974 total)
201 online now:
DrJones*, dwise1, JonF, Percy (Admin) (4 members, 197 visitors)
Newest Member: Howyoudo
Post Volume: Total: 875,775 Year: 7,523/23,288 Month: 82/1,347 Week: 99/342 Day: 26/56 Hour: 1/1


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Oracle Wins Ruling Against Google Over Java APIs
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 38 of 64 (849415)
03-08-2019 10:26 AM
Reply to: Message 36 by Hyroglyphx
03-08-2019 1:00 AM


Hyroglyphx writes:

Sure, why not?

You sound so confident, but the case between Google and Oracle has wended its way through the courts with each court reversing the one below. Google is currently appealing to the Supreme Court to rule on whether API's can be copyrighted.

But can I therefore co-opt the site's logo...

Google didn't do this. If you have an Android phone, try and find the Java logo.

...but its not an invitation to steal the product design.

But did Google steal Oracle's product design? If Azul can implement their own rangeCheck() function for Windows, why can't Google implement their own rangeCheck() function for Android? Is there an actual reason for the difference in Oracle's position concerning the two platforms, or is it just that Android and Google are where the money is?

--Percy


This message is a reply to:
 Message 36 by Hyroglyphx, posted 03-08-2019 1:00 AM Hyroglyphx has not yet responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 40 of 64 (849423)
03-08-2019 6:07 PM
Reply to: Message 39 by PaulK
03-08-2019 10:30 AM


Re: Oracle Further Monetizing Java
PaulK writes:

quote:

But it's all still frameworks for getting your Java-based product out to customers. I'm wondering why, if you're going to pay for something, you would buy it from Azul instead of Oracle? Is there some value added that Azul provides that Oracle doesn't? Is their price significantly lower (in which case can they still make a profit?)?


Likely they are cheaper - especially for those still sticking to Java 8. Oracle is not known for low prices.

But if Azul has made the investment to release their own versions of OpenJDK 9-12, how do they make money if they're mainly giving their software away while selling some enterprise versions of OpenJDK 8.

quote:

I don't know what "Zing JVM" is, but why would someone buy it from Azul instead of the equivalent from Oracle?

Given that Oracle’s JVM used to be free, I’m sure they have some selling points.

That's what I'm curious about - what are the selling points? Can Azul make money in the long term? They're private, so there's no annual financial info, just the amounts of VC funding.

quote:

You must be looking at a different webpage. Click on the link in the nested quotes above. Search for the string "11", which occurs in just one place and only in reference to the OpenJFX builds for OpenJDK 8 and 11. I don't know what OpenJFX is.

That isn’t the same link.

It's the link I was looking at when I wrote - I suspected you were looking at a different link, and that's I quoted the whole exchange for you.

But the title of the page is:

Download Zulu® tested, certified builds of OpenJDK

And I don’t understand how you could miss the fact that “OpenJFX for OpenJDK 8 and 11” explicitly refers to OpenJDK 11.

I didn't miss the fact. It talks about Azul's OpenJFX build for OpenJDK 8 and 11, but whose OpenJDK 8 and 11 they mean it doesn't say. Maybe it seems a stupid thing to be uncertain about, but I have mentioned several times that I'm not knowledgeable about all the vegetables in the Java garden. I don't even have analogs in my mind of what most of this stuff is. Does it help if I mention that I was an assembler/Fortran/Algol/SAIL/bliss/Pascal/c/C++ programmer who never used an IDE for development until I wrote an app in Swift a couple years ago? I still use emacs for most stuff.

(And if you actually click on the links you can see that OpenJDK 11 and 8 are available for download.)

I don't know what links you mean. There are no links for OpenJDK 8 and 11 at Download Zulu® tested, certified builds of OpenJDK or at OpenJFX link. The downloads on the latter page are for OpenJFX, not OpenJDK.

quote:

This is a prime example of why I find this so confusing, because another webpage talks about how JDK had to be reconfigured going forward to Java 9 but that they broke some things, implying (but not explicitly stating) that others (like Azul) wouldn't be able to implement their own versions beyond 8

But that doesn’t imply any such thing.

But that he reluctantly conceded that yeah, okay, they did break a few things, implies a problem. I've already said I'm coming at this from the outside with no help, that there are contradictory webpages, that I can't possibly know who to believe, and that I find the contradictory information confusing. But I am trying to understand it, and people don't make concessions like that unless there's a problem. So if the problem isn't that Azul and so forth would have trouble creating their own OpenJDK's beyond Java 8 then what's the problem?

OpenJDK is supported by Oracle and has the right to make the same changes.

I wasn't questioning whether Azul had the right but whether they had the resources.

--Percy


This message is a reply to:
 Message 39 by PaulK, posted 03-08-2019 10:30 AM PaulK has responded

Replies to this message:
 Message 41 by PaulK, posted 03-09-2019 1:59 AM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 42 of 64 (849425)
03-09-2019 10:27 AM
Reply to: Message 41 by PaulK
03-09-2019 1:59 AM


Re: Oracle Further Monetizing Java
I'm not sure how to thank someone for trying to explain something to me while insulting me at the same time. If you consider the Azul pages in isolation then it lessens the confusion, but if you include other pages I've linked to, like the one from the guy at Red Hat or the one from the guy at Oracle, they don't say the same thing as Azul. Who to believe?

Also, I think that sometimes you're looking at different webpages. I provided links to the pages I thought you were referring to, and they contained no links to OpenJDK builds. Zulu for Windows does contain such links, but this is the first time you linked to this page.

They don’t need the resources to write a complete JDK because they don’t do that.

But Oracle "broke" the JDK after 8. What did Azul, for example, do to address these problems, if anything? I don't know - do you? Can you take a Java project from Azul's OpenJDK 11 and have it work transparently on Oracle's OpenJDK 11? I don't know this either - do you?

My concern is that Oracle is going to find ways make such things more difficult, pushing more and more people toward purchasing solutions from Oracle while still trying to make it seem that they want to encourage rather than antagonize the larger Java community.

I think there's a comparison to be drawn with Red Hat, which took a different route. They didn't try to copyright APIs, and they didn't create convoluted library relationships that make it difficult for customers to know when they're using something free and when they're not.

--Percy


This message is a reply to:
 Message 41 by PaulK, posted 03-09-2019 1:59 AM PaulK has responded

Replies to this message:
 Message 43 by PaulK, posted 03-09-2019 11:27 AM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 44 of 64 (849446)
03-10-2019 12:23 PM
Reply to: Message 43 by PaulK
03-09-2019 11:27 AM


Re: Oracle Further Monetizing Java
PaulK writes:

quote:

I'm not sure how to thank someone for trying to explain something to me while insulting me at the same time.


If you find reporting what you are doing “insulting” then as I said, you have a problem

You're not reporting but contriving fictions. Who really has the problem, the person trying to figure things out, or the person saying, "Oh, I'll help," while telling them how idiotic they are? Do you know what the verb "to help" means? Or is it that you've got some rules you follow, that you help people who agree with you but invent ways to insult those who don't under the guise of help?

quote:

Also, I think that sometimes you're looking at different webpages. I provided links to the pages I thought you were referring to, and they contained no links to OpenJDK builds. Zulu for Windows does contain such links, but this is the first time you linked to this page.

But that page is one of the links on the Azul page that you originally provided. One of the prominent links that you couldn’t find. Despite the fact that it is very obvious.

But I wasn't looking for that link. I didn't link to that webpage because of its content but because it contained a count-up clock of how long it's been since the last free Oracle Java 8 update. I chose that page at random because many Azul pages contain that clock at the top.

Then you said, "The Azul page you link to mentions OpenJDK Java 11, too," but this isn't accurate. It does mention it, but only in the context of OpenJFX, and not as a link. I reported that back to you, and now you're trying to help me by telling me what an idiot I am for not noticing all the other links on the page. I don't need this kind of help. If this is how you're going to continue helping me then please just go away.

quote:

But Oracle "broke" the JDK after 8. What did Azul, for example, do to address these problems, if anything? I don't know - do you? Can you take a Java project from Azul's OpenJDK 11 and have it work transparently on Oracle's OpenJDK 11? I don't know this either - do you?

As I have said before they did not break the JDK - the changes they made to the JDK broke some older programs. And it’s for the OpenJDK project to address, not Azul.

But others say they did break JDK at Java 9, for instance see Java 9, it did break some things,' Oracle bod admits to devs still clinging to version 8, and that's just one of the links I came across that echo these sentiments. This is why I'm finding this all so confusing: there are claims that Oracle both did and did not break things, and that it does and it doesn't matter that they broke things.

quote:

My concern is that Oracle is going to find ways make such things more difficult, pushing more and more people toward purchasing solutions from Oracle while still trying to make it seem that they want to encourage rather than antagonize the larger Java community.

In which case the answer is for the community - users and developers - to embrace OpenJDK instead of paying Oracle.

This entails the risks I've been talking about, as summarized here in this optimistic piece The future of Java and OpenJDK updates without Oracle support:

quote:
Keeping Java updated in the absence of support from Oracle engineers will be a challenge to the Java community, but I believe it is one we should enthusiastically embrace. It is a golden opportunity for us, the community, to show what we can do. A truly open and transparent OpenJDK updates project will encourage wider participation and benefit all Java users.

He expresses confidence it can be done but says it represents a challenge. My own opinion is that despite making noises to the contrary, Oracle is a bad player in all this. They will try to ride a fine line that encourages the Java community while pushing as many as possible toward entrapment in their pay solutions. That many users don't even know they've stumbled over the boundary between free and pay content until notified by Oracle is pretty good evidence that this is so.

--Percy


This message is a reply to:
 Message 43 by PaulK, posted 03-09-2019 11:27 AM PaulK has responded

Replies to this message:
 Message 45 by PaulK, posted 03-10-2019 2:35 PM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 46 of 64 (849464)
03-10-2019 10:02 PM
Reply to: Message 45 by PaulK
03-10-2019 2:35 PM


Re: Oracle Further Monetizing Java
PaulK writes:

quote:

You're not reporting but contriving fictions. Who really has the problem, the person trying to figure things out, or the person saying, "Oh, I'll help," while telling them how idiotic they are? Do you know what the verb "to help" means? Or is it that you've got some rules you follow, that you help people who agree with you but invent ways to insult those who don't under the guise of help?


In reality you keep missing the obvious...

So when something obvious to you isn't obvious to someone else, this is what you do.

...and making excuses (sometimes bizarre) to avoid admitting it.

You're already guilty of making up your own version of the conversation, don't compound it.

And that is your problem. If you are insulted that I notice, that is also your problem.

That you're like this is your problem.

There really are no such links on that page:


I don't know what links you mean. There are no links for OpenJDK 8 and 11 at Download Zulu® tested, certified builds of OpenJDK

Despite the fact that the links are there and quite obvious. And just to make it more obvious, the page title is “Download Zulu® tested, certified builds of OpenJDK”.

If there are links to builds for OpenJDK on that webpage, then where are they? They're not there. I keep telling you this, but you keep coming back and telling me they are there. Where? Just quote the text of the sentence where the links are, that should do fine. Good luck.

Earlier I mentioned that it did mention OpenJDK 11 because it does. This is what it says:

Migrating from JavaFX? You can get free OpenJFX builds of Zulu for OpenJDK 8 and 11

Suddenly you've shifted from links to mentions, and you did this in earlier posts, too. And you frequently provide no links, making it impossible to be sure which webpage you're talking about. Anyway, yes, that's the sentence I'm talking about. You said the webpage contained links to builds for OpenJDK, but it doesn't, and especially that sentence doesn't. The "OpenJDK 8 and 11" part is just plain text, no links.

The actual links to webpages of links to builds are in this sentence:

quote:
Zulu builds of OpenJDK are available for Linux (including Alpine Linux with musl libc), Windows, macOS, Solaris, and Docker.

Now it does say that it is offering OpenJFX builds FOR OpenJDK 11 but that is certainly a mention of OpenJDK 11.

It most certainly is a mention, but not a link to a build of OpenJDK 11.

And do I really need to point out that it would be weird to offer something FOR software that doesn’t exist?

I'm not entirely sure how to interpret this sentence, but you do make it sound like something weird. But I never thought the software didn't exist. I at first was only sure it existed at Oracle for post Java 8 versions (and clearly expressed my uncertainty about whether it also existed elsewhere), and then later could see it existed at Azul, too. I never implied software was being offered that didn't exist (which is my guess at what your sentence actually means). That's just you making things up again.

In other words it as as I said. The JDK itself wasn’t broken - but the new JDK broke programs that had worked with earlier versions.

That's just a semantic distinction. "JDK's not broken, but the changes we made broke your program, so I guess you'll just have to fix it." What are the symptoms of broken programs? If it's just compilation failures then that's not too bad. If it's crashing until fixed then I guess that's not too bad, either. I hope it can never happen that their behavior can change, because that could be extremely difficult to find.

You won’t find me defending Oracle. But as I said the answer is to embrace OpenJDK. It has major backers who can fight for it (and have incentives to do so). It’s not easy for Oracle to trap you if you aren’t using their software.

I predict Oracle will make it harder and harder to avoid using anything from them. It's already in their DNA, as their history demonstrates.

--Percy


This message is a reply to:
 Message 45 by PaulK, posted 03-10-2019 2:35 PM PaulK has responded

Replies to this message:
 Message 47 by PaulK, posted 03-11-2019 1:49 AM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 48 of 64 (849473)
03-11-2019 12:05 PM
Reply to: Message 47 by PaulK
03-11-2019 1:49 AM


Re: Oracle Further Monetizing Java
PaulK writes:

quote:

So when something obvious to you isn't obvious to someone else, this is what you do.


No, it’s when they miss something that really is obvious - and start blaming other people and getting all snotty about it.

Ah, I see, if in your judgment somebody misses something really obvious then casting insults is fine and any attempts to set the record straight are met with more accusations. Good show. Perhaps you can tell us how being snotty is consistent with admissions of how confusing I was finding Java and how I didn't know what to think.

According to you you didn’t miss the reference to OpenJDK 11 on the Azul page. You just say - paraphrased - “maybe it means someone else’s OpenJDK 11” as if that would mean it wasn’t OpenJDK11.

Now you're just making stuff up again. You're correct that when first skimming the page I wasn't sure whose OpenJDK they were referring to in the context of OpenJFX, but of course the text said OpenJDK 8 and 11 (not just 11 as you said above) - it's right there in black and white. Why are you accusing me of making absurd misreadings I didn't make? And even if I did, so what? Can't you just say, "I think you missed where it said OpenJDK 8 and 11 - if you look at that part again it should make a lot more sense now."

What I did say was that I didn't know what OpenJFX was, and that following the link didn't tell me what it was, either. This just led to further uncertainty.

Or, more recently “it’s in the context of OpenJFX” as if that would mean it wasn’t a mention of OpenJDK 11.

I said I wasn't sure whose OpenJDK it was a reference to.

quote:

You're already guilty of making up your own version of the conversation, don't compound it

That’s a lie.

You just did it again above, so how could it be a lie?

quote:

That you're like this is your problem.

I’m sorry that you dislike honesty.

I do have an aversion to people making stuff up while declaring how honest they are.

quote:

There really are no such links on that page:

Really ? You can’t see the row of boxes across the page titled “Zulu for Windows”, “Zulu for Linux” and “Zulu for macOS” ? Each with a button labelled “Get Started” ?

Of course I can see these links, but what's Zulu? Is it just Azul's name for OpenJDK? Obviously it's in some way related to OpenJDK, but is it actually just a synonym for OpenJDK? When you kept saying there are links to OpenJDK builds right on that page did you mean the links that say Zulu?

Or the other links below, some of which are for OpenJDK builds for other platforms?

Can I assume you're referring to the links labeled Zulu?

quote:

Suddenly you've shifted from links to mentions, and you did this in earlier posts, too.

And there we have another lie. I was following your shift of subject.

And yet you can't quote me doing that, but I can quote you doing that. Here you are in just Message 41 alone:

If you are looking at a page titled:

Download Zulu® tested, certified builds of OpenJDK

and fail to find the links to download the OpenJDK builds you have a problem.

How about Zulu for Windows ?

If you can say

That part of the page doesn't actually say OpenJDK but OpenJFX,

while noticing that it explicitly mentions “OpenJDK 8 and 11” you have a problem (especially as the existence of OpenJDK 11 was the point in question)

So there you are switching from "links" to "mentions." I seriously assumed you were always talking about links. And now we know you were telling me I was missing OpenJDK links right before my very eyes when the links actually said Zulu.

We're spending a lot of time on Azul pages, but as I told you before, I was browsing through lots of webpages. I didn't have time to closely scrutinize each page as we've now done with this Azul page. Even the first time I read that page below the count-up clock portion I was just skimming for links to OpenJDK builds, meaning links that actually say "OpenJDK" - they're not there (there is a link to an early access build for OpenJDK 12, but I was looking for a regular release build).


But I wasn't looking for that link. I didn't link to that webpage because of its content but because it contained a count-up clock of how long it's been since the last free Oracle Java 8 update. I chose that page at random because many Azul pages contain that clock at the top.

Then you said, "The Azul page you link to mentions OpenJDK Java 11, too," but this isn't accurate. It does mention it, but only in the context of OpenJFX, and not as a link...

In the reality of the conversation I raised the issue of the mention first and never claimed it was a link Message 33

The context was links, so when you shifted to the word "mention" I assumed you were still talking about links, since who cares about a "mention" unless it's part of a paragraph that actually makes something clear. I think instead of all this abuse that a simple, "No, no, you've mistaken my meaning. I only meant plain text when I said mentions, not links," would have sufficed.

Instead of getting ridiculously pedantic...

Translated: "Just shut up and let me continue enumerating all your faults in peace."

...and insisting that links to the actual download pages don’t count maybe you should just admit that you didn’t notice those links in the first place.

Of course I didn't notice those links in the first place. My first link to that page was to quote their count-up clock. I already told you I picked that page at random because most of the Azul pages include that clock. I wasn't looking for links further down the page. Then you started telling me the page had links to OpenJDK builds, but again, there are no links to OpenJDK builds that actually say "OpenJDK".

quote:

It most certainly is a mention, but not a link to a build of OpenJDK 11.

So, saying that the page contains a mention of OpenJDK 11 IS accurate, despite your assertion to the contrary:


Then you said, "The Azul page you link to mentions OpenJDK Java 11, too," but this isn't accurate. It does mention it, but only in the context of OpenJFX, and not as a link

Obviously I said that last part before I realized that you weren't talking about mentions that were links - how is it not obvious to you now that I was thinking in the context of links, not plain text mentions. What you actually meant only became clear over time.

And let us not forget your initial reply:


That part of the page doesn't actually say OpenJDK but OpenJFX, and I don't know what OpenJFX is.

And I still don't know what OpenJFX is. I said what you quoted because OpenJFX is the only word in that sentence that's a link, and you were telling me to look for links, but to OpenJDK. So I was looking for the word OpenJDK as a link, but there isn't one. However the word OpenJDK does appear in the same sentence as OpenJFX as a link, so that's why I brought it up. I thought perhaps you had skimmed quickly and thought you saw an OpenJDK link when it was actually that OpenJFX link.

quote:

I'm not entirely sure how to interpret this sentence, but you do make it sound like something weird. But I never thought the software didn't exist. I at first was only sure it existed at Oracle for post Java 8 versions (and clearly expressed my uncertainty about whether it also existed elsewhere), and then later could see it existed at Azul, too. I never implied software was being offered that didn't exist (which is my guess at what your sentence actually means). That's just you making things up again.

No, I am not making things up. To restore context, you said:


What I'm finding is pretty confusing. Some webpages imply that OpenJDK isn't following Oracle to Java 9 and beyond, other webpages imply that it is. I don't know what to think.

And


I get the impression that it will take a concerted effort for OpenJDK to cross the hump from Java 8 to Java 9 and beyond, but like I said, some stuff on the web (like at java.net) implies they have crossed that hump. Again, I don't know what to think.

The fact that OpenJDK is up to Java 11 - the current version for Oracle - really does answer those points.

You're taking the absurd position that I should know things before I know them. I was pretty clear in expressing my uncertainty while I tried to understand Java better, with phrases you just quoted like, "What I'm finding is pretty confusing," and "Again, I don't know what to think."

quote:

That's just a semantic distinction. "JDK's not broken, but the changes we made broke your program, so I guess you'll just have to fix it."

No it isn’t. Oracle broke backward compatibility because that was the price for revamping Java so that it could go on. That is different from introducing bugs that make the JDK unusable. The later JDKs are not broken, as you claimed.

No matter how you cut it or what words you want to employ, breaking backward compatibility is a serious breach. It explains why Azul tries to emphasize the seriousness by including that count up clock about days since the last free Oracle Java 8 updates is on their almost every page.

quote:

If it's just compilation failures then that's not too bad. If it's crashing until fixed then I guess that's not too bad, either. I hope it can never happen that their behavior can change, because that could be extremely difficult to find.

It’s hard to say. Some of the changes were the removal of API calls which would just fail to compile. Whether there were more subtle changes as well I can’t say without researching it.

Sounds potentially serious. And they removed API calls? Seriously? I recall a similar occurrence when Red Hat discontinued support for the Motif graphic library maybe 15 years ago - it was hell. We had to reimplement key portions of important products.

quote:

I predict Oracle will make it harder and harder to avoid using anything from them. It's already in their DNA, as their history demonstrates.

They may try. They will not necessarily succeed.

Yep. We'll see.

--Percy


This message is a reply to:
 Message 47 by PaulK, posted 03-11-2019 1:49 AM PaulK has responded

Replies to this message:
 Message 49 by PaulK, posted 03-11-2019 1:53 PM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 50 of 64 (849516)
03-12-2019 9:28 PM
Reply to: Message 49 by PaulK
03-11-2019 1:53 PM


Re: Oracle Further Monetizing Java
Wow, aren't you just a bristling bundle of hostility.

PaulK writes:

quote:

Ah, I see, if in your judgment somebody misses something really obvious then casting insults is fine and any attempts to set the record straight are met with more accusations. Good show. Perhaps you can tell us how being snotty is consistent with admissions of how confusing I was finding Java and how I didn't know what to think.


By which you mean if I dare to point out that you are missing completely obvious things you will feel insulted. And if you “try to set the record straight” by inventing silly excuses it won’t work. And if you resort to lies and false accusations that won’t work either.

By which I mean that you've decided your words mean precisely what you say they mean and you deny any possible ambiguity or possibility they could be open to misinterpretation. It's almost like you're insisting that others make Talmudic analyses of what you say upon penalty of, well, this.

quote:

Now you're just making stuff up again. You're correct that when first skimming the page I wasn't sure whose OpenJDK they were referring to in the context of OpenJFX, but of course the text said OpenJDK 8 and 11 (not just 11 as you said above) - it's right there in black and white. Why are you accusing me of making absurd misreadings I didn't make? And even if I did, so what? Can't you just say, "I think you missed where it said OpenJDK 8 and 11 - if you look at that part again it should make a lot more sense now."

So you admit the substantive part...

That's not the substantive part.

...but try to cover it up with silly nitpicking.

What nitpicking?

I will point out that since you had no doubt that OpenJDK 8 existed there wasn’t really much point in emphasising it.

I will point out that when you said "OpenJDK 11" it caused me to search for a link on that page called "OpenJDK 11".

Now as to the substantive point someone else’s OpenJDK 11 (whatever you mean by that)...

How can you be critical of that whose meaning you say you do not understand. I think I explained this already, but in case not, I meant that I didn't know whether they meant their OpenJFX was built on their OpenJDK's or someone else's, like Oracle.

...would still be OpenJDK 11 so that isn’t even a relevant excuse.

You're ignoring what I've already told you that you find inconvenient. I thought I was being directed to links and there were no such links.

Not to mention the fact that you originally claimed that:


That part of the page doesn't actually say OpenJDK but OpenJFX...


I said I wasn't sure whose OpenJDK it was a reference to.

And later you said


Then you said, "The Azul page you link to mentions OpenJDK Java 11, too," but this isn't accurate. It does mention it, but only in the context of OpenJFX, and not as a link.

Odd how “in the context of OpenJFX, and not as a link” somehow means it is inaccurate to say that the page “mentions OpenJDK 11”

Odd how you can't understand that since you had claimed there were links to OpenJDK 11 and I had said I couldn't find them that I still thought you were talking about links to OpenJDK.

quote:

I do have an aversion to people making stuff up while declaring how honest they are.

You also clearly have an aversion to people pointing out your obvious errors and silly excuses. Because that is what is happening here.

You also clearly have an aversion to people pointing out your obvious lack of clarity. Because that is what's been happening.

quote:

Of course I can see these links, but what's Zulu? Is it just Azul's name for OpenJDK? Obviously it's in some way related to OpenJDK, but is it actually just a synonym for OpenJDK? When you kept saying there are links to OpenJDK builds right on that page did you mean the links that say Zulu?

And there you go missing the obvious again. The title of the page is:


Download Zulu® tested, certified builds of OpenJDK

So “Zulu” is their trademarked name for their builds of OpenJDK

You're seeing what you want to see. What it actually says is that you can download builds of OpenJDK that were tested and certified by Zulu, and at the time I had no idea that Zulu was Azul's name for OpenJDK. Once you know that Zulu is a synonym for OpenJDK it's easy to see what they really meant.

quote:

And yet you can't quote me doing that

Which is a really odd thing to say when I did quote you doing that,

Here it is again:


But I wasn't looking for that link. I didn't link to that webpage because of its content but because it contained a count-up clock of how long it's been since the last free Oracle Java 8 update. I chose that page at random because many Azul pages contain that clock at the top.

Then you said, "The Azul page you link to mentions OpenJDK Java 11, too," but this isn't accurate. It does mention it, but only in the context of OpenJFX, and not as a link...

quote:

So there you are switching from "links" to "mentions." I seriously assumed you were always talking about links. And now we know you were telling me I was missing OpenJDK links right before my very eyes when the links actually said Zulu.

Let us note that your trouble with the mention started before I started talking about the links so that excuse doesn’t wash.

You are very confused.

And yes I did expect you to read the title and realise that Azul called their “tested, certified builds of OpenJDK” Zulu.

Except it doesn't say that, as I explained above.

quote:

The context was links, so when you shifted to the word "mention" I assumed you were still talking about links, since who cares about a "mention" unless it's part of a paragraph that actually makes something clear. I think instead of all this abuse that a simple, "No, no, you've mistaken my meaning. I only meant plain text when I said mentions, not links," would have sufficed.

The first few lines of my message are:


There were some major revisions with Java 9 that required some code rewrites.

OpenJDK is currently supported up to Java 11 and Java 12 is coming (9 and 10 are out of support). The Azul page you link to mentions OpenJDK Java 11, too.

Now if you had actually followed the link I did give - to the OpenJDK’s project you would have found links to OpenJDK 11. But apparently you decided that a “mention” had to be a link instead.

Good grief, here we go again. There is no link to OpenJDK 11 on that page (just to be excruciating clear, the OpenJDK java.net webpage), at least not one using the text "OpenJDK 11". It does mention Oracle's stuff, but I already knew that existed.

You've been constantly wrong about where you claimed links were. And could you at least include a link to the message you're quoting from instead of making me search for your Message 33? You're terrible about providing links.

And I am sure that if I had pointed out that a “mention” would naturally be text you would be insulted.

I never said I was insulted. I said you were being insulting. Insult away since you apparently can't help it. And at the risk of pushing you further over the edge into petulance, peevishness and pique, I'll repeat that your writing is less than clear, except apparently to you.

quote:

Of course I didn't notice those links in the first place. My first link to that page was to quote their count-up clock. I already told you I picked that page at random because most of the Azul pages include that clock. I wasn't looking for links further down the page. Then you started telling me the page had links to OpenJDK builds, but again, there are no links to OpenJDK builds that actually say "OpenJDK".

Because obviously a page titled:


Download Zulu® tested, certified builds of OpenJDK

Wouldn’t have links to download pages for OpenJDK builds ?

Been over this already. There are no links to OpenJDK builds on that page that are labeled OpenJDK.

quote:

And I still don't know what OpenJFX is. I said what you quoted because OpenJFX is the only word in that sentence that's a link, and you were telling me to look for links, but to OpenJDK

In the post where I brought it up - quoted above - I didn’t tell you to look for links and you had already conceded there was a mention before I did.

Again, you are very confused.

quote:

And I still don't know what OpenJFX is.

And I don’t either. Feel free to research it yourself.

In the context of someone newly acquainting himself with Java, and with the confused information you were providing, how was I to know whether OpenJFX was important or not. I'm guessing not right now.

quote:

I said what you quoted because OpenJFX is the only word in that sentence that's a link, and you were telling me to look for links, but to OpenJDK. So I was looking for the word OpenJDK as a link, but there isn't one. However the word OpenJDK does appear in the same sentence as OpenJFX as a link, so that's why I brought it up. I thought perhaps you had skimmed quickly and thought you saw an OpenJDK link when it was actually that OpenJFX link.

Except, of course I didn’t tell you to look for links, and “it doesn’t actually say” hardly means “it doesn’t link to”.

Except, of course, you told me there were links. You said there were links to OpenJDK 11 right on that page, except of course that there aren't, not by the name "OpenJDK 11".

quote:

You're taking the absurd position that I should know things before I know them. I was pretty clear in expressing my uncertainty while I tried to understand Java better, with phrases you just quoted like, "What I'm finding is pretty confusing," and "Again, I don't know what to think."

Except that I didn’t. I provided evidence that OpenJDK was up to Java 11 - and you managed to get into a mess over the idea that “OpenJDK 8 or 11” is a direct reference to OpenJDK 11.

What's important about the words you use is not the meaning they convey to your own mind, but the meaning they convey to others. If you're happy that your meaning is clear only to yourself then that isn't very helpful to anyone else.

You never commented on the OpenJDK page at all.

Again, there is no link to OpenJDK 11 (other than Oracle's) on that page. I never commented on the page because it wasn't helpful.

quote:

No matter how you cut it or what words you want to employ, breaking backward compatibility is a serious breach. It explains why Azul tries to emphasize the seriousness by including that count up clock about days since the last free Oracle Java 8 updates is on their almost every page.

The question is not whether it is a serious breach. The question is whether the OpenJDK project had to do anything to “fix” the breaks. And - so far as OpenJDK itself is concerned - the answer is clearly no. That is why it matters that the JDK itself was not broken, as you claimed.

If the new release breaks your code, then does the fact that they did it on purpose mean their new release isn't broken? But they're Oracle and everyone else isn't, so of course they're right. Right?

And why, given my incipient level of understanding of things Java, would it be deserving of all this disapprobation you're pouring upon me to initially think that those releasing their own OpenJDK's wouldn't attempt remedies?

You might find this hard to believe, but people could fail to follow your meaning without being grievously at fault.

--Percy


This message is a reply to:
 Message 49 by PaulK, posted 03-11-2019 1:53 PM PaulK has responded

Replies to this message:
 Message 51 by PaulK, posted 03-13-2019 2:23 AM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 52 of 64 (849557)
03-14-2019 1:10 PM
Reply to: Message 51 by PaulK
03-13-2019 2:23 AM


Re: Oracle Further Monetizing Java
PaulK writes:

quote:

By which I mean that you've decided your words mean precisely what you say they mean and you deny any possible ambiguity or possibility they could be open to misinterpretation. It's almost like you're insisting that others make Talmudic analyses of what you say upon penalty of, well, this.


As can be seen in your example all I am asking for is simple English comprehension. The label”Talmudic” is better applied to your attempts to hide your mistakes.

Paraphrasing, "No, I'm not Talmudic, you are?" Really?

quote:

That's not the substantive part.

The substantive part is that your assertion was false.

Except that it was true.


That part of the page doesn't actually say OpenJDK but OpenJFX,...

It does say OpenJDK - and you claimed to know that it did say OpenJDK. The rest is just excuses which make no sense.

You are again very confused.

quote:

I will point out that when you said "OpenJDK 11" it caused me to search for a link on that page called "OpenJDK 11".

However I never mentioned a link,...

I thought that to you "mentions" and "links" were two different things, yet here you are "mentioning a link." Amazing. And it isn't true you never mentioned a link - of course you did. For example, this is an excerpt from my Message 37:

Percy in Message 37 writes:

PaulK writes:

Percy writes:

PaulK writes:

Percy writes:

For example, the top of this webpage at Azul Systems has a count-up clock of how long it's been since the last free Oracle Java 8 update: Download Zulu® tested, certified builds of OpenJDK.

The Azul page you link to mentions OpenJDK Java 11, too.

That part of the page doesn't actually say OpenJDK but OpenJFX,...

That certainly isn’t what I see. I see the home page for OpenJDK, with OpenJFX only one link near the bottom of a very long list of links.

You must be looking at a different webpage. Click on the link in the nested quotes above. Search for the string "11", which occurs in just one place and only in reference to the OpenJFX builds for OpenJDK 8 and 11. I don't know what OpenJFX is.

[Channel Paulk]
How could you make such a basic mistake? There's the link to the webpage referred to sitting right there in the nested quote where I said it was (Download Zulu® tested, certified builds of OpenJDK), and if you go to that webpage and search for the string "11" just as I instructed you'll be brought to a sentence with a link to OpenJFX. This is basic. How could you screw it up and then just make a bunch of transparent excuses for it? (Although I did just notice that if the count-up clock happens to have an "11" or two in it that the "11" in that sentence will no longer be the first "11" on the page but the second or third.)
[/Channel Paulk]

...nor did I ask you to look for links earlier in that post - nor did you say that there was a mention but not a link. You outright claimed that it did not say OpenJDK.

You're either very confused or purposefully misrepresenting.

quote:

How can you be critical of that whose meaning you say you do not understand. I think I explained this already, but in case not, I meant that I didn't know whether they meant their OpenJFX was built on their OpenJDK's or someone else's, like Oracle.

Let’s guess. You are confusing a specific JDK - OpenJDK - with JDKs in general.

No, and especially not on that page. Of course you provide no link again, so we could easily be talking about different webpages, so I guess it's up to me to remove the ambiguity: Download Zulu® tested, certified builds of OpenJDK

However OpenJDK is the JDK produced by the OpenJDK project. Indeed only a few posts earlier I introduced OpenJDK as a free alternative to Oracle’s software. The point at issue was that OpenJDK had got to Java 11 because you claimed to find conflicting information about it. So if you did make that confusion it could only be by ignoring the context of the conversation - but without that confusion it makes no sense.

Even ignoring that you provided no links to the messages you referenced, this is indecipherable

Because OpenJDK is an open source JDK owned by the OpenJDK project other people like Azul and Oracle can supply builds of it but they are all OpenJDK - and Azul meant all of them.

Am I supposed to read your mind about what Azul page you're talking about? These responses are what I mean by you constantly thinking you're a beacon of clarity when you're really not.

quote:

You're ignoring what I've already told you that you find inconvenient. I thought I was being directed to links and there were no such links.

However, as I have pointed out neither I nor you made any mention of links until later in the conversation. Even in your initial defence - that you quoted - you don’t make the claim that you are looking for links. And since your supposed reason - my mention of the links on the Azul page - had yet to be written I am left with no reason to think that you were actually looking for links then.

If you're not going to link to the messages you're talking about, this is again indecipherable. I can only say it sounds wrong and that you may be mischaracterizing the conversation.

quote:

Odd how you can't understand that since you had claimed there were links to OpenJDK 11 and I had said I couldn't find them that I still thought you were talking about links to OpenJDK.

Aside from the fact that that can only possibly apply to the third quote the real oddity is that you didn’t notice that what you said contradicted itself.

Which "third quote?" Of me? Of you? In this message? In some other message? You're being incredibly obscurantist without even trying.

quote:

You also clearly have an aversion to people pointing out your obvious lack of clarity. Because that is what's been happening.

Funny how you haven’t cited a single example then.

I've cited plenty, and plenty more so far in this post alone. Get your head out of the sand. Your lack of clarity is frequent.

quote:

You're seeing what you want to see. What it actually says is that you can download builds of OpenJDK that were tested and certified by Zulu, and at the time I had no idea that Zulu was Azul's name for OpenJDK. Once you know that Zulu is a synonym for OpenJDK it's easy to see what they really meant.

And you can quite easily work out that “Zulu” is their name for their builds of OpenJDK.

Right under the title it says:


Zulu is Free to Download and Use.

You wouldn’t download a certifying organisation.

And where does it say that Zulu is a "certifying organization"? Nowhere, and I didn't think that. I had no idea what Zulu was. I put it in the category, "This doesn't look important, file away to figure out later."

Even if you imagine that Zulu is a tool for certifying builds of OpenJDK you should quickly realise that all the downloads offered are Zulu or Zulu related. Since the page offers downloads of OpenJDK it doesn’t take much of a leap to realise that Zulu should be OpenJDK.

Someone truly interested in helping someone else get a clear picture of things would have quickly recognized the Zulu/OpenJDK confusion.

And finally if you can’t find links that I say are prominent - and which should be there - would it not be a whole lot better to simply say so rather than declaring that there are no such links?

I did say so. I said there's no such link on that page.

quote:

You are very confused.

Because I think that message 33 was posted before message 39 (where I first mentioned the links on the Azul page) ?

You're not going to type "[msg=33]" instead of "message 33"? You're not going to include your own words that I called confused so I can know what you're responding to? You're very consistent in your lack of clarity.

quote:

Except it doesn't say that, as I explained above.

I didn’t claim that it said that, but I do claim that it can be quite easily worked out and I explain just how to do that above.

You quote me saying, "Except it doesn't say that," but you don't quote what you said that I was responding to. How can I know what this is about unless you quote that? Again, complete lack of clarity. Faith was very similar in her freedom with pronouns. You have to substitute the actual nouns back in from time to time otherwise your meaning becomes completely lost.

quote:

Good grief, here we go again. There is no link to OpenJDK 11 on that page (just to be excruciating clear, the OpenJDK java.net webpage), at least not one using the text "OpenJDK 11". It does mention Oracle's stuff, but I already knew that existed.

It says:


Download and install the open-source JDK for most popular Linux distributions. Oracle's OpenJDK JDK 11 binaries are at jdk.java.net/11; Oracle's JDK 11 product binaries for Linux, macOS, and Windows, based largely on the same code, are here.

The first link is to builds of OpenJDK 11, just as I said. The fact that Oracle supplied the builds does not change the fact that they are builds of OpenJDK and not Oracle’s own JDK - as the text says, Oracle’s own JDK is at the second link.

Well now you're just being dunderheaded. Those are just links to Oracle's stuff. How did you miss where I said, " It does mention Oracle's stuff, but I already knew that existed." You quoted it just above. Look up about an inch, there it is.

Also, your way of describing links would lead almost anyone astray. When you tell people a certain webpage contains a link to "X", they're not going to look for a link called "Y".

quote:

You've been constantly wrong about where you claimed links were.

By which you mean that you are been constantly failing to find links which are there, exactly where I said.

You're fooling no one but yourself. The links as you described them don't exist.

I guess that is what you mean by “being insulting”. But it is still just telling the truth.

Self-serving, yes, truth, no.

quote:

Been over this already. There are no links to OpenJDK builds on that page that are labeled OpenJDK.

But you have yet to address the point that the title of the page says that the links are there, that providing those links is the purpose of the page. That should be enough to understand that the links are there, despite the labelling.

The web abounds with misleading webpages. I assumed you were giving me accurate information that the link was called "OpenJDK 11", but you weren't. When I didn't find a link called "OpenJDK 11" I told you. I definitely didn't start clicking on every link on the webpage hoping one would take me to the OpenJDK 11 build.

quote:

Except, of course, you told me there were links. You said there were links to OpenJDK 11 right on that page, except of course that there aren't, not by the name "OpenJDK 11".

By which you mean that there are obvious links but you weren’t able to identify them.

You're being purposefully obtuse. As I've already said, why, before all this discussion, would anyone think a link labeled "Zulu" was a link to "OpenJDK"?

quote:

In the context of someone newly acquainting himself with Java, and with the confused information you were providing, how was I to know whether OpenJFX was important or not. I'm guessing not right now.

In the context of a discussion about OpenJDK I would think that OpenJDK would be more important than software offered for OpenJDK releases. And OpenJFX is clearly the latter.

I thought you said you didn't know what OpenJFX is, but now you know that it's "software offered for OpenJDK releases." Interesting how the story changes.

quote:

What's important about the words you use is not the meaning they convey to your own mind, but the meaning they convey to others. If you're happy that your meaning is clear only to yourself then that isn't very helpful to anyone else.

I’m confident that it is clear to people that aren’t confusing themselves.

Ah, a legend in your own mind, but all the evidence says you aren't very clear. You rarely say what webpages or messages you're referring to, and you rarely describe links accurately. The examples abound in this post alone.

quote:

Again, there is no link to OpenJDK 11 (other than Oracle's) on that page. I never commented on the page because it wasn't helpful.

In other words there is a link to OpenJDK 11 but you dismissed it because you saw the word “Oracle” without bothering to understand that OpenJDK isn’t Oracle’s JDK despite the context of the conversation and despite the fact that the distinction is present in that very paragraph.

Oracle’s build of OpenJDK 11 is still OpenJDK 11 and proof that OpenJDK 11 exists.

You sure don't read very carefully. As I've said multiple times across multiple messages now, I already knew Oracle had OpenJDK's beyond Java 8. You are very confused.

quote:

If the new release breaks your code, then does the fact that they did it on purpose mean their new release isn't broken? But they're Oracle and everyone else isn't, so of course they're right. Right?

The fact that Oracle owns and defines the Java language and the JDK definition does mean that they get to change them for new releases, and are allowed to break backward compatibility. Obviously.

Sure, they're allowed to do whatever they like, but that makes Oracle a bad citizen of the Java community. An honest and open citizen of that community would make Java an open standard, say by giving responsibility to ANSI and ISO.

It also means that anyone implementing to the same specification doesn’t have to fix the breakages, which is the actual point under discussion.

Actually, it's not "the actual point under discussion." I already acknowledged that I now understand that it would mean violating the Oracle-defined standard.

quote:

And why, given my incipient level of understanding of things Java, would it be deserving of all this disapprobation you're pouring upon me to initially think that those releasing their own OpenJDK's wouldn't attempt remedies?

That would be their own JDKs. Their JDKs have to meet Oracle specifications and that is pretty much all they have to do. They don’t have to add “remedies” and they probably can’t if those remedies would come into conflict with Oracle’s changes to the spec.

Yes, I understand that now. The question is why you think I should have understood that from the beginning, and why you think explaining that I couldn't possibly know what I do not yet know is making excuses.

quote:

You might find this hard to believe, but people could fail to follow your meaning without being grievously at fault.

I didn’t say that trying to win an argument by changing the subject -even inadvertently - was a grievous fault. But it is still a fault. “Azul have to do a lot of work” is different from “Oracle did a bad thing”.

I think you're very confused again. Repeating what I said another way, it takes two to communicate information, one to send and one to receive. When the message fails to be clearly communicated one cannot automatically assume the receiver is at fault, which is what you seem determined to prove. Why is it so important to you that whatever's gone wrong has been all my fault? Why does it even have to be anyone's fault? People misunderstand each other the first or second time incredibly frequently - it doesn't mean someone involved must be a bad player.

Providing some feedback: a) When quoting from a message, provide enough so that the context can be understood; something like "That's not the substantive part," doesn't provide anything to go on and forces people to have to read back to old messages; the quoting facility is provided to make that unnecessary; b) When referencing old messages, provide a link - the dBCode is fewer characters anyway and is easier, unless you have a real aversion to typing square brackets; c) When referring to webpages don't say things like "that webpage" - provide a link; it is a bit more work, but only a bit with copy/paste; d) When referring to a link on a webpage, use the actual text of the link.

Of course you don't have to do any of this, but if you don't then don't blame others when they don't find your posts as crystal clear as you do.

--Percy


This message is a reply to:
 Message 51 by PaulK, posted 03-13-2019 2:23 AM PaulK has responded

Replies to this message:
 Message 53 by PaulK, posted 03-14-2019 3:43 PM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


(1)
Message 54 of 64 (849621)
03-16-2019 12:13 PM
Reply to: Message 53 by PaulK
03-14-2019 3:43 PM


Re: Oracle Further Monetizing Java
Thanks for your help.

--Percy


This message is a reply to:
 Message 53 by PaulK, posted 03-14-2019 3:43 PM PaulK has not yet responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 55 of 64 (849748)
03-20-2019 7:40 AM


Supreme Court Urged to Overturn Lower Court API Ruling
Summarizing how this thread began, Google won a jury ruling that their use of Oracle's Java API's was fair use. Oracle appealed to the Federal Circuit and got the ruling overturned. I predicted that Google would appeal to the Supreme Court and that many amicus curiae briefs would be filed. This has come true, as reported in an article that appeared last month but that I only just noticed: US Supremes urged by pretty much everyone in software dev to probe Oracle's 'disastrous' Java API copyright win

Google has appealed to the Supreme Court, and 14 amicus briefs have been filed, among them by Microsoft, Red Hat, Mozilla, the Python Software Foundation, Developers Alliance and the EFF (Electronic Frontier Foundation). Excerpts:

quote:
"The federal circuit's decision threatens disastrous consequences for innovation," said Microsoft (PDF). "[It] threatens fair use's vitality and extinguishes the necessary 'breathing room' for the ecosystem of innovation it protects."

A group submission from Mozilla, Mapbox, Medium, Patreon, Etsy and Wikimedia said that the court had "heedlessly unraveled" a "reasonably predictive rule" and a set of "reliable norms" that are critical for coders to understand what is and is not appropriate to carry over from one project to another.

"The decisions call into question a bedrock legal assumption that has fostered tremendous innovation," said Red Hat (PDF). "[The decisions] threaten disruption across technology and communications industries including individual developers, small companies, and companies of substantial size.

The Developers Alliance also sought to emphasize the knock-on effects of the decisions. “The current case has implications that go far beyond the two litigants involved,” it said (PDF). “In 2017 there were an estimated three million software developers in the United States, and their collective work added an estimated $565 billion to the country's gross domestic product."
...
A group of 78 computer scientists said in their filing (PDF) that, by extending copyright protection to interfaces, the court showed a "fundamental misunderstanding" of how interfaces differ from programs.

Software interfaces, such as those embodied in the Java API, they said, "are purely functional systems or methods of operating a computer program or platform," not programs themselves.

"Interfaces merely describe what functional tasks a computer program will perform without specifying how it does so," they said. In contrast, implementations provide the actual step-by-step instructions to perform each task included in an interface."


--Percy

Edited by Percy, : Fix up character misrenderings.


  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 57 of 64 (849929)
03-26-2019 8:31 AM


Concern HIgh Over Oracle's "Gotcha" Methods
A recent article in Business Insider says that Oracle customers fear Oracle's gotcha methods when it comes to running their software in non-Oracle clouds such as AWS (Amazon Web Services, but basically their cloud) or Azure (Microsoft's cloud). Oracle has made all the same reassuring sounds that they've made with Java, but customers remain concerned. A few excerpts:

quote:
Customers fear that choosing a cloud that competes with Oracle's cloud may encourage Oracle to try and fine them for contract violations.
...
Large enterprises that use Oracle software know they want to move that software to the cloud but have some fears about Oracle's reaction if they don't choose its cloud, according to a new survey of 300 IT professionals.
...
There are good reasons for companies to move slowly when they're shifting these systems to the cloud.
...
But the top concern companies had for dragging their feet, according to the survey, was a fear that if they chose an Oracle competitor like Amazon's AWS to be their cloud provider, they would trigger an audit from Oracle.

The dreaded audit is when Oracle examines how a company is using its software. If Oracle finds that the company has violated its software license contract, Oracle can slam its customer with big fees and other threats, like terminating permission to use its software.

And the terms in Oracle's contract, like other big enterprise vendors, are notoriously complicated. In fact, there's an entire consulting industry called software asset management that helps companies negotiate software contracts and fight audits and audit fees from Oracle and others (including Microsoft, SAP, IBM). [In other words, Oracle isn't the only bad citizen out there. -Percy]
...
Even so, in this survey, 58% said they were very or somewhat worried that their licenses will not remain valid if they moved to a non-Oracle cloud.
...
But there's a big caveat: there may be hidden terms in their Oracle contract that would allow Oracle to audit them.
...
One of the most famous Oracle software asset management consultants says Oracle deliberately makes its contracts vague and difficult which gives it the upper hand.

Craig Guarente, CEO of Palisade Compliance used to be an Oracle exec who conducted audits and now makes his living helping Oracle customers defend themselves against them.

Oracle has recently launched a campaign promising to cut customers bills in half if they switch from AWS to Oracle's cloud.

But Guarente calls the campaign a trap, he wrote in a recent blog post.


Of course use of Java would be scrutinized along with everything else under an Oracle audit, and as time goes by use of OpenJDK will likely provide less and less safety. Many libraries were invalidated at Java 9, and they're only slowly attaining post Java 9 compatibility. Some are finding that compliance isn't possible while staying on the OpenJDK side of the fence.

I have no doubt that Oracle is working on non-OpenJDK features and capabilities (and licensing) that they hope will leave those not using them on the outside looking in. OpenJDK will still work fine, but since it can't use (for a made up example) the proprietary Oracle Super Cloud Analytics Library it will be much less desirable as part of someone's cloud-based solution.

There's an example of what Java could be right at this website: PHP. No company, public or private, owns PHP. It's "owned" by the users. There is no formal specification (this isn't necessarily a good thing and one is under development), just a reference implementation (this is "the code is the standard" approach). Changes to PHP are in the direction that users want it to go. Java should be the same way, an open standard driven by users, not by Oracle.

About "the code is the standard" approach, it can have some serious downsides. This was never demonstrated more clearly than in a language few will likely have heard of, Verilog, a language for describing digital hardware like that in your computer, cell phone or TV. Practically every digital device you use was designed using Verilog.

When Verilog first became a public standard (through the IEEE, it's IEEE standard 1364) there was both a specification (from the IEEE) and a de facto reference implementation (from the company that donated Verilog to the public domain). The reference implementation wasn't perfect, but because it was already very widely used competitors actually advertised that their implementations were bug-for-bug compatible with the reference implementation, even though there was a specification. Too many users of the reference implementation were able to successfully get their chips out the door, and they wanted no changes.

This situation gradually resolved itself over time (the bugs were fixed a good 20 years ago), but it's a lesson in how "the code is the standard" is not a good idea.

That reminds me. I was limited on the old server to PHP 5, but now that we're in the cloud I can go to PHP 7, which I plan to do soon since PHP 5 could go away at some point. This will entail some MySQL changes - the old MySQL API I'm using was first deprecated and finally discontinued in PHP 7.

--Percy


Replies to this message:
 Message 58 by PaulK, posted 03-26-2019 1:52 PM Percy has acknowledged this reply

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 59 of 64 (861292)
08-19-2019 12:38 PM


Changes Are Already Having an Impact
Last year when I used New Hampshire's online registry search to find a deed it was simple and easy. I am just at this minute trying to use it again and the webpage has this notice:

**NOTICE: Oracle changed their terms of use in Java 8 update 211. Please uninstall this version and install Java 8 update 201 which would have been the version you were using before updating. This version keeps you in compliance with Oracle's terms of use licensing and allows the search application to load.

When I installed the specified Java update it provided this notice:

Important Information about Oracle Java SE Roadmap

Changes are coming which will impact your access to future releases of Java SE from Oracle

Corporate users will be impacted with the April 2019 update.

These changes do not affect the version you are about to install.

For additional guidance please follow the link below.

More information...

One obvious concern is how long old versions of Java will continue to work. If the deed registry changes their software they can currently produce Java 8 update 201, but will they still be able to produce that release in, say, 2022? Will they be required to purchase a license or change to something other than Java? What if I need to use other software that uses update 211. I'll have to install it, but then switch back to update 201 the next time I need to look up a deed. Naturally when I last posted I was in no position to anticipate specific negative impacts of Oracle's licensing changes, but this is precisely the kind of thing I meant by being at the mercy of Oracle. I should be happily going about my day by now, but instead I'm retro-installing an old version of Java and trying to figure out why the webpage still doesn't work.

--Percy


Replies to this message:
 Message 61 by RAZD, posted 08-27-2019 5:02 PM Percy has responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 60 of 64 (861825)
08-27-2019 1:31 PM


Oracle Gives Itself Another Black Eye
Reinforcing the impression that they're an untrustworthy company, Oracle is reneging on the "free for life" service promised by acquired company Dyn. The "free for life" promise was made to those who made donations to Dyn in its early days, providing it essential funding. Oracle is offering these Dyn customers its Oracle Cloud product, claiming it is very inexpensive.

To support this claim they cite their cheapest possible option, $0.85/million-queries. To get an idea how much that is, this website gets about 100M queries/month. A former Dyn customer with a website about as active as this one would pay $85/month, which is very reasonable but far more than the $0 they were paying before in return for having donated to Dyn. This is apart from any additional services they might have received from Dyn, such as higher bandwidth or disk space or control panels, etc.

Source: Oracle customers cause a Dyn over withdrawal of lifetime licences

--Percy


  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


(1)
Message 62 of 64 (861843)
08-27-2019 8:04 PM
Reply to: Message 61 by RAZD
08-27-2019 5:02 PM


Re: Changes Are Already Having an Impact
RAZD writes:

You could set up a similar box for Java 201.

Thanks for the suggestion, VirtualBox sounds neat, but it turns out that once I got Java 201 going that I was able to fix my registry problem. I shouldn't need to use New Hampshire's online real estate registry again.

But the problem with Oracle and Java remains. What happened to me virtually out of the blue shows that Oracle's Java licensing actions are not innocuous. They are not something your average person doesn't have to worry about. There must be many other Java software programs caught in the same bind, and we have to be concerned about how fast the set of Java programs that no longer run on the currently released version will grow, affecting yet more people.

Oracle apologists need to relearn the tale of the camel's nose. Oracle is using the carrot of free Java to lure increasing numbers of unsuspecting users into the grasp of their non-free Java. They're a good corporate citizen the way Trump is a very stable genius. This isn't an argument against corporations making profits, just against underhanded sneakiness, which is easy to hide in the software world. Volkswagen provides a good example. No one objects to Volkswagen making money on their diesel vehicles, but rigging the software to adjust how the engine runs when connected to emissions testers is beyond the pale. What Oracle is doing is different only stylistically - it's still dishonest.

--Percy


This message is a reply to:
 Message 61 by RAZD, posted 08-27-2019 5:02 PM RAZD has not yet responded

  
Percy
Member
Posts: 19801
From: New Hampshire
Joined: 12-23-2000
Member Rating: 2.4


Message 63 of 64 (866789)
11-15-2019 9:51 PM


Supreme Court Agrees to Hear Google’s Appeal
The Supreme Court has agreed to hear Google’s appeal in its legal battle with Oracle over whether software API interface definitions can be copyrighted: Supreme Court will hear Google's appeal in massive copyright suit brought by Oracle

—Percy


Replies to this message:
 Message 64 by Percy, posted 11-16-2019 7:55 AM Percy has acknowledged this reply

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2018 by EvC Forum, All Rights Reserved

™ Version 4.0 Beta
Innovative software from Qwixotic © 2020