Gotta learn enough to be able to make original contributions
Well, I had a procedure on the Monday of Thanksgiving to fix an issue with my nose being blocked that I've had since I was a kid, so I spent the week recuperating from that at my parents' new house. Now I am glad to be home at my apartment, but sad to do work.
Yep. Probably not a lot of room for creativity, at least in finding discoveries that actually work.
Ah, very interesting. Glad you're feeling a little better and hopefully breathing a lot clearer too. It'll feel nice having some fresh oxygen to do your work with.
However, for some good innovations I guess a little creativity is necessary.
Yes, it's been quite strange getting used to the changes. For a while I was confused every time I looked in the mirror.
I actually found a way to circumvent almost all of ProBoards' development rules (in a way that PB can't prohibit). PB doesn't allow AJAX, iframes, or requests to the server or database (besides with very limited ways).
BUT, you can make a get or post request or use an iframe or just do stuff with non-PB servers or databases. So I created a web server that can track TBB in certain ways and then do things. For example, I programmed something yesterday that tracks the last active threads that a user has, and then puts that information on the TBB profile page to make it that the profile page has the user's last three topics (like on LMB's profile pages). Without the workaround (put the information from TBB on my external website, and then get that information from the external website onto TBB with AJAX) that was previously impossible. (btw, those three last threads hasn't been added to TBB yet) I could also do something like track if a specific page is loaded, and then make it that a "views" counter is added to - each time a specific page is loaded. I could track how long a person is on TBB and then make that a factor when it comes to rank. I could add a live chat to TBB and make it that when a user posts in the live chat it adds the live chat message to the user's posts list (/user/id/recent). I could make it that on any click or any load or anything really, something is added to a database, and then I could put that info on TBB. Honestly, I'm probably not explaining it in a good way, but like, just know that I figured out how to do a lot of stuff.
I also figured out how to add an AI (which I would train with something like tensorflow.js or brain.js, and it could make a post hidden and flagged if the AI labels it as inappropriate), but that AI doesn't have much to do with my "external server and database" workaround.
I had to read that slowly but I understand what you are saying. So you can implement a lot of things... but is it worth the bother? Is the epic MB simulation worth pursuing?
Also, never expected to see someone talking about tensorflow on the lego message boards, but here we are. Comes up a lot in ML math talks... now I understand why people complain about the computer scientists implementing things without knowing the underlying theory ( )
But hey man, if you want to do this stuff, you could make it into a pretty impressive portfolio. Don't downplay the things you do here... unlike me you're doing more substantial things than vomiting words
I actually found a way to circumvent almost all of ProBoards' development rules (in a way that PB can't prohibit). PB doesn't allow AJAX, iframes, or requests to the server or database (besides with very limited ways).
BUT, you can make a get or post request or use an iframe or just do stuff with non-PB servers or databases. So I created a web server that can track TBB in certain ways and then do things. For example, I programmed something yesterday that tracks the last active threads that a user has, and then puts that information on the TBB profile page to make it that the profile page has the user's last three topics (like on LMB's profile pages). Without the workaround (put the information from TBB on my external website, and then get that information from the external website onto TBB with AJAX) that was previously impossible. (btw, those three last threads hasn't been added to TBB yet) I could also do something like track if a specific page is loaded, and then make it that a "views" counter is added to - each time a specific page is loaded. I could track how long a person is on TBB and then make that a factor when it comes to rank. I could add a live chat to TBB and make it that when a user posts in the live chat it adds the live chat message to the user's posts list (/user/id/recent). I could make it that on any click or any load or anything really, something is added to a database, and then I could put that info on TBB. Honestly, I'm probably not explaining it in a good way, but like, just know that I figured out how to do a lot of stuff.
I also figured out how to add an AI (which I would train with something like tensorflow.js or brain.js, and it could make a post hidden and flagged if the AI labels it as inappropriate), but that AI doesn't have much to do with my "external server and database" workaround.
I had to read that slowly but I understand what you are saying. So you can implement a lot of things... but is it worth the bother? Is the epic MB simulation worth pursuing?
Also, never expected to see someone talking about tensorflow on the lego message boards, but here we are. Comes up a lot in ML math talks... now I understand why people complain about the computer scientists implementing things without knowing the underlying theory ( )
But hey man, if you want to do this stuff, you could make it into a pretty impressive portfolio. Don't downplay the things you do here... unlike me you're doing more substantial things than vomiting words
Yeah, that's the problem with all things TBB-dev-related. I'm 90% done with an Avatar Maker for this website (that uses the LEGO.com avatar parts before LEGO life), and I have tons of sub-projects for this website, but I just hope there's enough people to use the features, or else they're not really worth it. Like, a big chunk of the twenty people who visit everyday use a ProBoards theme instead of our LMB theme, so is it worth supporting a nostalgic masterpiece if there isn't enough people to use it? (By the way, I'm planning on making the LMB theme more convenient (like the LMBE themes) and far more mobile friendly) But alas, I have an activity plan which might help, and once (or rather, if) we have more people, the features may be worth the work. Also, features could help with activity, since convenience and accuracy both make this site more appealing to potential members and inactive members.
Yes. I can't comprehend how hard it must have been for the creators of tensorflow to make it, but it's the same thing with so many computer science things. For example, no one looks at C developers and says "Wow you have no idea what's really going on! You don't know how hard the creators of C worked to compile C code to assembly. You have no idea of the lexical processes and how much torture it must have been for the dev's to make a programming language that you can actually use.. How ignorant of you." I think it's kind of the same thing. Like yeah, tensorflow is a great thing for people who don't want to get their hands dirty in the complex topic that is artificial intelligence and machine learning, just like C is a great thing for people who don't want to torture themselves with assembly. You know what I'm saying?
I had to read that slowly but I understand what you are saying. So you can implement a lot of things... but is it worth the bother? Is the epic MB simulation worth pursuing?
Also, never expected to see someone talking about tensorflow on the lego message boards, but here we are. Comes up a lot in ML math talks... now I understand why people complain about the computer scientists implementing things without knowing the underlying theory ( )
But hey man, if you want to do this stuff, you could make it into a pretty impressive portfolio. Don't downplay the things you do here... unlike me you're doing more substantial things than vomiting words
Yeah, that's the problem with all things TBB-dev-related. I'm 90% done with an Avatar Maker for this website (that uses the LEGO.com avatar parts before LEGO life), and I have tons of sub-projects for this website, but I just hope there's enough people to use the features, or else they're not really worth it. Like, a big chunk of the twenty people who visit everyday use a ProBoards theme instead of our LMB theme, so is it worth supporting a nostalgic masterpiece if there isn't enough people to use it? (By the way, I'm planning on making the LMB theme more convenient (like the LMBE themes) and far more mobile friendly) But alas, I have an activity plan which might help, and once (or rather, if) we have more people, the features may be worth the work. Also, features could help with activity, since convenience and accuracy both make this site more appealing to potential members and inactive members.
Yes. I can't comprehend how hard it must have been for the creators of tensorflow to make it, but it's the same thing with so many computer science things. For example, no one looks at C developers and says "Wow you have no idea what's really going on! You don't know how hard the creators of C worked to compile C code to assembly. You have no idea of the lexical processes and how much torture it must have been for the dev's to make a programming language that you can actually use.. How ignorant of you." I think it's kind of the same thing. Like yeah, tensorflow is a great thing for people who don't want to get their hands dirty in the complex topic that is artificial intelligence and machine learning, just like C is a great thing for people who don't want to torture themselves with assembly. You know what I'm saying?
I mean, I'd use the LMB theme if it was actually the real mccoy... I remain skeptical if you can actually increase activity here, and if activity did ramp up I'd probably leave... enough to catch up on as is.
Fair point. My career goal / interest is in creating more efficient and applicable ML algorithms so perhaps you can pardon my bemusement.
Yeah, that's the problem with all things TBB-dev-related. I'm 90% done with an Avatar Maker for this website (that uses the LEGO.com avatar parts before LEGO life), and I have tons of sub-projects for this website, but I just hope there's enough people to use the features, or else they're not really worth it. Like, a big chunk of the twenty people who visit everyday use a ProBoards theme instead of our LMB theme, so is it worth supporting a nostalgic masterpiece if there isn't enough people to use it? (By the way, I'm planning on making the LMB theme more convenient (like the LMBE themes) and far more mobile friendly) But alas, I have an activity plan which might help, and once (or rather, if) we have more people, the features may be worth the work. Also, features could help with activity, since convenience and accuracy both make this site more appealing to potential members and inactive members.
Yes. I can't comprehend how hard it must have been for the creators of tensorflow to make it, but it's the same thing with so many computer science things. For example, no one looks at C developers and says "Wow you have no idea what's really going on! You don't know how hard the creators of C worked to compile C code to assembly. You have no idea of the lexical processes and how much torture it must have been for the dev's to make a programming language that you can actually use.. How ignorant of you." I think it's kind of the same thing. Like yeah, tensorflow is a great thing for people who don't want to get their hands dirty in the complex topic that is artificial intelligence and machine learning, just like C is a great thing for people who don't want to torture themselves with assembly. You know what I'm saying?
I mean, I'd use the LMB theme if it was actually the real mccoy... I remain skeptical if you can actually increase activity here, and if activity did ramp up I'd probably leave... enough to catch up on as is.
Fair point. My career goal / interest is in creating more efficient and applicable ML algorithms so perhaps you can pardon my bemusement.
What could I do to make it the "real McCoy"? Oh no, it's an ultimatum.
Ah, that sounds interesting. What types of ML models have you made?
I mean, I'd use the LMB theme if it was actually the real mccoy... I remain skeptical if you can actually increase activity here, and if activity did ramp up I'd probably leave... enough to catch up on as is.
Fair point. My career goal / interest is in creating more efficient and applicable ML algorithms so perhaps you can pardon my bemusement.
What could I do to make it the "real McCoy"? Oh no, it's an ultimatum.
Ah, that sounds interesting. What types of ML models have you made?
Well, it doesn't look exactly like the MBs, so I'd rather use something nothing like the MBs than a cobbled together hybrid. Also, if you got studs to work I'd probably use the old MB theme for nostalgia
And yes it's the benboy paradox
Well, I'm not quite sure I'd go with "made" but I've learned about and written simple code for linear SVM (regression and classification), linear discriminant analysis, ridge regression, and my professor had a strange fixation on ADMM so we used that a lot. Maybe it's used a lot in practice too, who knows. I need to practice coding some neural nets and decision trees so I can understand the underlying machinery.
What could I do to make it the "real McCoy"? Oh no, it's an ultimatum.
Ah, that sounds interesting. What types of ML models have you made?
Well, it doesn't look exactly like the MBs, so I'd rather use something nothing like the MBs than a cobbled together hybrid. Also, if you got studs to work I'd probably use the old MB theme for nostalgia
And yes it's the benboy paradox
Well, I'm not quite sure I'd go with "made" but I've learned about and written simple code for linear SVM (regression and classification), linear discriminant analysis, ridge regression, and my professor had a strange fixation on ADMM so we used that a lot. Maybe it's used a lot in practice too, who knows. I need to practice coding some neural nets and decision trees so I can understand the underlying machinery.
What about it doesn't look like the MBs? I'd say it's pretty accurate, (although the profile page is in progress).
I've already asked Sam if I can remake the old MB theme, but he said no, even though I could add studs and make it perfect.
Well, it doesn't look exactly like the MBs, so I'd rather use something nothing like the MBs than a cobbled together hybrid. Also, if you got studs to work I'd probably use the old MB theme for nostalgia
And yes it's the benboy paradox
Well, I'm not quite sure I'd go with "made" but I've learned about and written simple code for linear SVM (regression and classification), linear discriminant analysis, ridge regression, and my professor had a strange fixation on ADMM so we used that a lot. Maybe it's used a lot in practice too, who knows. I need to practice coding some neural nets and decision trees so I can understand the underlying machinery.
What about it doesn't look like the MBs? I'd say it's pretty accurate, (although the profile page is in progress).
I've already asked Sam if I can remake the old MB theme, but he said no, even though I could add studs and make it perfect.
The profile pages, the leaderboard pages, the likes being a convoluted mess...
Why didn't he want you doing that? Why does it matter?
Well, it doesn't look exactly like the MBs, so I'd rather use something nothing like the MBs than a cobbled together hybrid. Also, if you got studs to work I'd probably use the old MB theme for nostalgia
And yes it's the benboy paradox
Well, I'm not quite sure I'd go with "made" but I've learned about and written simple code for linear SVM (regression and classification), linear discriminant analysis, ridge regression, and my professor had a strange fixation on ADMM so we used that a lot. Maybe it's used a lot in practice too, who knows. I need to practice coding some neural nets and decision trees so I can understand the underlying machinery.
What about it doesn't look like the MBs? I'd say it's pretty accurate, (although the profile page is in progress).
I've already asked Sam if I can remake the old MB theme, but he said no, even though I could add studs and make it perfect.
Now it just crossed my mind, could you make the ranks so they depend on a bunch of things.. like "X posts or Y likes" That'd be fun.
What about it doesn't look like the MBs? I'd say it's pretty accurate, (although the profile page is in progress).
I've already asked Sam if I can remake the old MB theme, but he said no, even though I could add studs and make it perfect.
The profile pages, the leaderboard pages, the likes being a convoluted mess...
Why didn't he want you doing that? Why does it matter?
The profile page is getting a complete makeover (I'm almost done with the accurate version of it), and I've been trying to fix the likes system (and I'm making progress). What exactly do you mean by the leaderboard pages? Also, in response to your other post: Yes, making ranks depend on posts and likes is possible, and I suggested it years ago, but no one on staff is really in favor of revamping the whole system.
We wanted to dedicate all our TBB development time to the LMB Theme and not the Old LMB theme that only a few people were interested in.
The profile pages, the leaderboard pages, the likes being a convoluted mess...
Why didn't he want you doing that? Why does it matter?
The profile page is getting a complete makeover (I'm almost done with the accurate version of it), and I've been trying to fix the likes system (and I'm making progress). What exactly do you mean by the leaderboard pages? Also, in response to your other post: Yes, making ranks depend on posts and likes is possible, and I suggested it years ago, but no one on staff is really in favor of revamping the whole system.
We wanted to dedicate all our TBB development time to the LMB Theme and not the Old LMB theme that only a few people were interested in.
It doesn't LOOK and FEEL like the MBs henny
Apparently this means I should just buy out LMBE so it can be my MBs fantasy...
Listen man, only a few people are interested in the LMB Theme
The profile page is getting a complete makeover (I'm almost done with the accurate version of it), and I've been trying to fix the likes system (and I'm making progress). What exactly do you mean by the leaderboard pages? Also, in response to your other post: Yes, making ranks depend on posts and likes is possible, and I suggested it years ago, but no one on staff is really in favor of revamping the whole system.
We wanted to dedicate all our TBB development time to the LMB Theme and not the Old LMB theme that only a few people were interested in.
It doesn't LOOK and FEEL like the MBs henny
Apparently this means I should just buy out LMBE so it can be my MBs fantasy...
Listen man, only a few people are interested in the LMB Theme
Okay, but what changes need to be made for it to "feel" like the MBs?
Yes, buy out LMBE.
I'm not the one you should say that to. I'm the one who agrees with you and wanted to make the old LMB Theme.
Apparently this means I should just buy out LMBE so it can be my MBs fantasy...
Listen man, only a few people are interested in the LMB Theme
Okay, but what changes need to be made for it to "feel" like the MBs?
Yes, buy out LMBE.
I'm not the one you should say that to. I'm the one who agrees with you and wanted to make the old LMB Theme.
I told you, it's the profile pages, the likes pages, the users online... all the weird little things add up for me. Furthermore, the lack of real lego avatars is a buzzkill. Speaking of that I do have all the ORIGINAL MB avatar heads (like 220 of them) saved if you could make use of that.
Maybe I should join staff. Are you still accepting new members?
![[fp]](http://storage.proboards.com/6672327/images/nnJcspnhTDEdUFpOzKkD.png)
Goodbye LMBs...