ubuntu 22.04 make error

[mjsduncan]

after changing CMakeLists.txt to point to directly to /usr/include/tbb, cmake is successful but hits an error in the tbb library (package version 2021.5.0-7ubuntu2)

In file included from /home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.cc:48:
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.h: At global scope:
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.h:61:38: error: expected class-name before ‘{’ token
   61 | class lambda_task : public tbb::task {
      |                                      ^
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.h:63:10: error: ‘task’ in namespace ‘tbb’ does not name a type
   63 |     tbb::task* execute() {
      |          ^~~~
In file included from /usr/include/tbb/task.h:17,
                 from /home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.cc:31:
/usr/include/oneapi/tbb/task.h:26:11: note: ‘tbb::v1::task’ declared here
   26 | namespace task {
      |           ^~~~
In file included from /home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.cc:48:
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.h: In function ‘void tbb_enqueue_lambda(const F&)’:
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.h:73:16: error: ‘enqueue’ is not a member of ‘tbb::v1::task’
   73 |     tbb::task::enqueue(*new(tbb::task::allocate_root()) lambda_task<F>(f));
      |                ^~~~~~~
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.h:73:40: error: ‘allocate_root’ is not a member of ‘tbb::v1::task’
   73 |     tbb::task::enqueue(*new(tbb::task::allocate_root()) lambda_task<F>(f));
      |                                        ^~~~~~~~~~~~~
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.cc: In member function ‘Json::Value opencog::AtomSpacePublisherModule::atomToJSON(opencog::Handle)’:
/home/mozi/oc/atomspace-restful/opencog/events/AtomSpacePublisherModule.cc:196:40: error: cannot convert ‘std::back_insert_iterator<std::vector<opencog::Handle> >’ to ‘const opencog::AtomSpace*’
  196 |         h->getIncomingSet(back_inserter(incomingHandles));
      |                           ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
      |                                        |
      |                                        std::back_insert_iterator<std::vector<opencog::Handle> >

[linas]

Quick hack is to change tbb::task to tbb::v1::task and maybe you'll get lucky. There might be more, after this one.

Correct long term is to port atomspace-visualizer to not use the restful interfaces. I mean, even if you fix this, you still have the other problem that the restful API is just sending wrong/bad data.

[linas]

"Doctor Doctor it hurts when I do this" -- "Well don't do that!"

It's time to bite the bullet, and pursue one of the options outlined in opencog/atomspace-explorer#8 The code in this git repo needs to be nuked.

[linas]

Sorry to spam you, Mike, but -- even if you get it to compile, you then have to solve the swagger issue, to make it run. Even if you solve the swagger issue, then you will get python errors when you run it. Even if you solve the python errors, you still have the problem that it is sending incorrect & obsolete data. The grand-total effort needed to fix this trainwreck of problems is larger than the amount of effort to port the atomspace explorer to working interfaces.

[mjsduncan]

i know for a fact this isn't true because eyob got it working quite easily in ubuntu 20.04. however obsolete you think json is it's ubiquitous and just fixing this is the quickest way to getting useful visualizer, lack of which is a key reason opencog has never gained traction

…

On Tue, May 17, 2022 at 4:52 PM Linas Vepštas ***@***.***> wrote: Closed #4 <#4>. — Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABGJSQQM43OVFE7YTA337CDVKQBI7ANCNFSM5WF55S4A> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[linas]

Mike, don't say things with aren't true, which then lead to arguments and hurt feelings. I'm trying to steer you in the right direction. The code here is crap. Give up on it, and do the right thing.

I mean -- ask yourself, why am I giving you this advice? What do you think my motivation is, for saying what I'm telling you? You think I'm ... what ... a used-car salesman trying to trick you into ... what, exactly? Buying a brand-new car? The code here is crap. It's been crap since forever. Stop investing time and effort into crap.

I mean, look, I can give you full write access to this repo (if you don't already have it), and you can do whatever the heck you want to do with it. Really, I just don't care. Have at it. Have fun! I'm just telling you, its a broke-down pile of rubble, but if that's how you want to live, .... whatever. I'm just giving you honest advice. I'm not sure why we have to argue about it.

[mjsduncan]

i don't know what you think is untrue, i've seen vid of opencog explorer running through atomspace-restful with eyob's tweaks. the tbb library has been totally rewritten so more tweaks are needed to make it ubuntu current. it remains the only decent atomspace visualizer i've ever used, what is the alternative? you are shutting down development of perfectly useable code with no viable replacement because you want someone to make something your way. i'm not a developer, i just want to visualize hypergraphs so i can experiment with knowledge representation. i'm not trying to argue, i'm pointing out that your gatekeeping of the opencog codebase drives away interested potential users.

…

On Tue, May 17, 2022 at 5:39 PM Linas Vepštas ***@***.***> wrote: Mike, don't say things with aren't true, which then lead to arguments and hurt feelings. I'm trying to steer you in the right direction. The code here is crap. Give up on it, and do the right thing. I mean -- ask yourself, why am I giving you this advice? What do you think my motivation is, for saying what I'm telling you? You think I'm ... what ... a used-car salesman trying to trick you into ... what, exactly? Buying a brand-new car? The code here is crap. It's been crap since forever. Stop investing time and effort into crap. I mean, look, I can give you full write access to this repo (if you don't already have it), and you can do whatever the heck you want to do with it. Really, I just don't care. Have at it. Have fun! I'm just telling you, its a broke-down pile of rubble, but if that's how you want to live, .... whatever. I'm just giving you honest advice. I'm not sure why we have to argue about it. — Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABGJSQQOLVFKMUKA226A6CDVKQGYHANCNFSM5WF55S4A> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[linas]

I don't know what you think is untrue

It's untrue, because I too have attempted to run this code. I have spent more than a few days on trying to make it work. I've tried to make this work, keep it up to date, for a decade. I know that it does not work, for me, and if I can't make it work, then probably most people cannot make it work. The fact that @Eyob was able to make it work .. well, bravo! He should share the secret sauce, the magic incantation for making it work.

you are shutting down development of perfectly useable code

It is not "perfectly usable", its a turdball wrapped in a cowpie in a field of fertilizer. You can read the source code, and see that it stinks. You can read the API, and see that it's wrong. You try to compile it yourself, and see that it doesn't compile.

But if you think it is usable, then go ahead and use it!

with no viable replacement

Kricky! I spell out three viable replacements right here: opencog/atomspace-explorer#8

you want someone to make something your way.

WTF. Then don't do it my way! Do it some other way! I'm trying to guide you to the simplest, easiest, least painful solution. And for this, you accuse me of wanting "someone to make something my way"?? Seriously? Don't take my advice if you don't want it! But don't ask for my help, and then when I offer it, get rude and insulting about it.

your gatekeeping of the opencog codebase drives away interested potential users.

That's 100% bullshit. You've repeatedly stated this b.s to my face over the years, and repeating it over and over does not make it true.

I've already given you full write permission to this repo. I'll give anyone else you want full write permission to do whatever you please with this repo. There are about a dozen people who have this admin authority, including Ben, Cassio, Amen, Nil, all the usual suspects.

I can create a brand-new repo for you, if that is what you want. Again, there are a dozen other people who can do this, as well. I'm not stopping you.

[mjsduncan]

all you have to do is keep the issue open

[mjsduncan]

thank you for keeping it open. maybe someone from the rocca crew will get around to pushing their working code. i'm still getting a build error & am giving up, but i did notice the rocca repo specifies atomspace commit b15826224 so maybe that has something to do with why you can't get it to build either.

> Kricky! I spell out three viable replacements right here:

opencog/atomspace-explorer#8 <opencog/atomspace-explorer#8> those are coding projects, not replacements.

> I know that it does not work, *for me*, and if I can't make it work,

then probably most people cannot make it work. ...and then you shut down. this is exactly what i mean by "gatekeeping"

…

On Tue, May 17, 2022 at 7:00 PM Linas Vepštas ***@***.***> wrote: Reopened #4 <#4>. — Reply to this email directly, view it on GitHub <#4 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABGJSQUQAMRXIU6GSDCVQ4LVKQQKTANCNFSM5WF55S4A> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[linas]

Well, let me know if there's something I can do, or what I could say or do differently.