Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Starting to Demo the Wolfram Language [video]

blog.stephenwolfram.com

Post reply on HN
Print view View original

31–40 of 75 posts

Re: Starting to Demo the Wolfram Language [video]

#31

Post by anigbrowl »

jjoonathan wrote: post #23
anigbrowl wrote: post #11

Earlier quoted context omitted.

What happens if you run it on itself - eg you do some sort of basic program with a little bit of structure and generate code from it, and then analyze the code, eg to make a graph of it (if that makes sense). Or to put it another way, what are the possibilities of making better tools Mathematica (or Wolfram Language or whatever it is called now) using Mathematica?

It would be slightly easier than coding Mathematica from scratch (you have better computers and an example to follow) but much harder to fund, since you would be re-treading old ground and wouldn't be able to competitive for a long time. The types of problems simplified by Mathematica don't include coding Mathematica. To be sure, it pioneered a few nifty techniques that you would want to take advantage of, especially…

Thanks for the comprehensive answer! I phrased the question poorly, in that I didn't intend to suggest running Mathmatica against its own codebase but rather against shorter end-user programs with a view to developing IDE-type addons.

I agree with your general comments, as I hate writing glue code and my interest is primarily for DSP and simulation applications, for which I already admire it.

Re: Starting to Demo the Wolfram Language [video]

#32

Post by beachstartup »

one use case could be a business analytics tool. 'business intelligence'.

i.e., import 30 days of log data and instantly plot the geo-located hits on a world map, as a time series of dynamically rendered graphical points, with built-in information i.e. size, color, icon, shape, etc. and also statistical analysis to go with it.

this scenario and many others are certainly possible today but would require quite a bit of code, libraries, and external api calls, as well as a wrapper such as a web app if you wanted a large audience to view it.

i think in the future, managers/executives who will separate themselves from the MBA herd are the ones who will be able to effectively use tools such as this.

Re: Starting to Demo the Wolfram Language [video]

#33

Post by dkural »

This is really fantastic of course, but the video doesn't answer the following question: The video says it’s the most powerful language, one can build so much etc. What’s one of those things, that they've built that no one else could build before? What great application now exists that didn’t before? I will still use google for search, twitter for opinion (that’s where the people are at..), photoshop for photo manipulation etc. Besides doing mathematics, which Mathematica is great at, what concrete evidence is there that this language is better for solving any particular problem one would actually care to solve? I don’t care about dominant flag colors.

Re: Starting to Demo the Wolfram Language [video]

#34

Post by jjoonathan »

anigbrowl wrote: post #31
jjoonathan wrote: post #23

Earlier quoted context omitted.

It would be slightly easier than coding Mathematica from scratch (you have better computers and an example to follow) but much harder to fund, since you would be re-treading old ground and wouldn't be able to competitive for a long time. The types of problems simplified by Mathematica don't include coding Mathematica. To be sure, it pioneered a few nifty techniques that you would want to take advantage of, especially…

Thanks for the comprehensive answer! I phrased the question poorly, in that I didn't intend to suggest running Mathmatica against its own codebase but rather against shorter end-user programs with a view to developing IDE-type addons. I agree with your general comments, as I hate writing glue code and my interest is primarily for DSP and simulation applications, for which I already admire it.

Whoops, sounds like you already know exactly what I spent 3/4 of my post belaboring :)

I'm still not sure I understand your proposal. How do short IDE-type addons solve the problem of integrating siloed codebases with each other? The way I see it, the only way forward is

1. Convince a language person that first-class support for algebraic metaprogramming is important

2. Convince someone from each of the fields {numerical methods, linear programming, finite element methods, graph theory, etc} to export an interface to the language in #1, possibly re-structuring their entire codebase to do so

3. Documentation, convenience schleps

I say "convince" because the problem of making a Mathematica replacement is far enough down my todo list that I have no choice but to be realistic about the fact that I probably won't be that person.

Re: Starting to Demo the Wolfram Language [video]

#35

Post by dnautics »

sadfaceunread wrote: post #20
dnautics wrote: post #18

Earlier quoted context omitted.

IIRC the stealth biotech startup "emerald therapeutics" is using mathematica to automate everything. I actually am personally biased against this startup, but it is a fascinating use case. EDIT: https://www.wolfram.com/mathematica/customer-stories/researc...

Really interesting use case. I'm super impressed with the concept, of automated data generation and gathering and having a common platform for the company. Perhaps the most impressive idea is that the data that is generated appears to be programatically accessible. I haven't worked in a large enterprise that has advanced past unsearchable and poorly indexed excel files that are manually manipulated/updated floating a…

For these guys the 'hard stuff' that really impresses me is interfacing with hardware. Scientific instrument controllers are notoriously proprietary and annoying to deal with. Though truth be told, in a small scientific environment not having user permissions is a 'feature not bug', especially if it's not siloed against other departments in, say a bigger facility.

Aside from personally knowing the CEO from grad school, a former co-worker of mine got hired into Emerald and basically transformed from a bioinformatician to a programming lackey... He may have hated every minute of it and quit.

Re: Starting to Demo the Wolfram Language [video]

#37

Post by nl »

I'm working on something akin to this, although my version is somewhat more ambitious[1].

So far I have a "Knowledge Engine" that can answer (some) questions in natural language by using a process similar to how IBM Watson Deep QA works (an extremely simplified version atm, though).

I'm currently working on integrating SymPy for symbolic calculation[2]. SymPy already has a web interface[3] which is broadly similar to what WolframAlpha can do. That relies on Google AppEngine to provide security sandboxing.

My version is designed to be completely private, open source and self-hostable, including all the data[1], so I can't rely on AppEngine. Instead, I have SymPy running inside ZeroVm + LXC (Docker) as a sandbox. I'm currently working out how to get that to communicate with the front end (which is non-trivial, because ZeroVm is pretty much undocumented).

Obviously SymPy isn't as "Natural Language Like" as the WolframAlpha language, but arguably Python is actually more powerful.

After that I'll do OpenStreetMap integration (ie, self hosted OSM), which will give it better geographic understanding that it has.

[1] Every time I mention this I need to include a standard disclaimer that I do realize this is completely and utterly crazy.

[2] http://sympy.org/en/index.html

[3] http://live.sympy.org/

Re: Starting to Demo the Wolfram Language [video]

#38

Post by taliesinb »

primitivesuave wrote: post #27
tedsanders wrote: post #15

Earlier quoted context omitted.

Could you give some more detail on how it failed to power Wolfram Alpha? Did they switch to an alternate language?

Mathematica for handling data aggregation, recursive descent parsing, and tasks well suited for it, while the rest is an enormous codebase of compiled and JVM languages stitched together with Mathematica. Such projects are referred to as the "mud bowl", where when something breaks you just have to throw more mud at it.

> recursive descent parsing

The parser is Java, actually. And it's not recursive descent.

> tasks well suited for it

Fancy that! Using a language for things it is well suited to! :)

> compiled and JVM languages stitched together with Mathematica

A lot of stuff is still WL though. Data curation, evaluation, scanning, typesetting, formatting, and, you know, the actual computation in doing something like http://www.wolframalpha.com/input/?i=regression+of+literacy%...

But it seems like you're complaining the backend isn't written in Wolfram Language? The JVM is great at that stuff! I don't call that ball-of-mud at all, just using the right tool for the job.

Re: Starting to Demo the Wolfram Language [video]

#39

Post by taliesinb »

anigbrowl wrote: post #11
primitivesuave wrote: post #6

I worked on Mathematica for 2 years, and really the biggest downside is how its an incredibly powerful tool that fails in many practical applications. It failed to power the backend of a search engine (Wolfram Alpha), and based on my latest version of Mathematica, the dynamic computing features, visualization, and JIT compilation features still have a long way to go. It's an incredible and fascinating tool when you'r…

What happens if you run it on itself - eg you do some sort of basic program with a little bit of structure and generate code from it, and then analyze the code, eg to make a graph of it (if that makes sense). Or to put it another way, what are the possibilities of making better tools Mathematica (or Wolfram Language or whatever it is called now) using Mathematica?

Yes! The prospect of doing that deeply and seriously is probably the single thing I'm most excited about tackling next.

There's a whole tier of previously impractical metaprogramming that would come with being able to analyze and manipulate programs (and commit histories) as easily as data.

Syntactic macros are just the teeniest tip of the iceberg!

Re: Starting to Demo the Wolfram Language [video]

#40

Post by taliesinb »

nl wrote: post #37

I'm working on something akin to this, although my version is somewhat more ambitious[1]. So far I have a "Knowledge Engine" that can answer (some) questions in natural language by using a process similar to how IBM Watson Deep QA works (an extremely simplified version atm, though). I'm currently working on integrating SymPy for symbolic calculation[2]. SymPy already has a web interface[3] which is broadly similar to…

How do you represent queries? That's really, for me, the most interesting question.
Post reply on HN