Wolfram Compute Services | Hacker News Reader

Wolfram Compute Services

writings.stephenwolfram.com

230 pointsnsoonhui136 comments

Man, I miss Wolfram Language. Once you've twisted your brain a little to grok its usage, it's such an incredibly high-value tool, especially for exploration and prototyping. I saw it more as a do-anything software tool for researchers rather than as a language aimed at programmers, so I put on a researcher hat and tried to forget everything I knew as a professional programmer, and had a few memorable seasons with it around 2016-2020. I remember calculating precisely which days of the year would cause the sunlight to pass through a window and some glass blocks in an internal wall, creating a beautiful light show indoors. It only took a couple of minutes to get a nice animated visualisation and a calendar.

Nowadays I'd probably just ask Claude to figure it out for me, but pre LLMs, WL was the highest value tool for thought in my toolbox.

(Edit: and they actually offer perpetual licenses!)

The power of the language came from the concise syntax (I liked it more then classical LISPs) with the huge library of Mathematica. When Python is "batteries included", Mathematica is "spaceship included".

If this was open sourced, it had the potential to severely change the software/IT industry. As an expensive proprietary software however, it is deemed to stay a niche product mainly for academia.

That's exactly the same analogy I used to use, although I said "nuclear reactor included" - spaceship is better, it implies less danger and more expanded horizons!

> As an expensive proprietary software however

It's $195/year for a personal license. And only $75/year for students. Their licensing model is pretty broad.

It still creates a class of haves and have-nots which prevents forming a community.

Plus you buy a version of it, and then someone else is on another version, and you don't have the same features, and the tiny community is fragmented.

Well, that doesn't sound too bad. But this is a high enough barrier for Mathematica to not see wide spread use.

I don't remember what the pricing has been throughout the years. But I do remember that for some of the time I couldn't really afford Mathematica. And the license I wanted was also a bit too expensive to justify for a piece of software that only I would be using within an organization.

Because it is also about enough other people around you not being able to justify the expense. And about companies not wanting to pay a lot of money for licenses so they can lock their computations into an ecosystem that is very small.

Mathematica is, in the computing world, pretty irrelevant. And I'm being generous when I say "pretty": I have never encountered it in any job or even in academia. People know of it. They just don't use it for work.

It would have been nice if the language and the runtime had been open source. But Wolfram didn't want to go in that direction. That's a perfectly fine choice to make. But it does mean that as a language, Mathematica will never be important. Nor will knowing how to program in it be a marketable skill.

(To Stephen Wolfram it really doesn't matter. He obviously makes a good living. I'm not sure I'd bother with the noise and stress coming from open sourcing something)

> And I'm being generous when I say "pretty": I have never encountered it in any job or even in academia. People know of it. They just don't use it for work.

To my knowledge, at least in academia, Wolfram (Mathematica) seems to be used quite a bit by physicists. Also in some areas of mathematics it is used (but many mathematicians seems to prefer Maple). Concerning mathematical research, I want to mention that by now also some open-source (and often more specialized) CASs seem to have become more widespread, such as SageMath, SymPy, Macaulay2, GP/PARI or GAP.

In Maple sin(x) is "sin(x)", in Mathematica it's "Sin[x]", ewww

In my opinion, Wolfram/Mathematica is more consistent internally, while Maple is more consistent with the usual mathematical notation.

> while Maple is more consistent with the usual mathematical notation

I can't tell if you're saying that as if it's a good thing, or a bad thing.

It's not about good nor bad, but about the different trade-offs that these two CASs made. What is more important for you is something that you can only answer for yourself.

I actually loved this idea so much that every language I make, I try to do the same. The point of it is that typing ( requires shift, while [ does not. And you have no idea when you have tunnel syndrome, how much it hurts each time you write a (. While it’s ugly, the hand thanks you for it.

> The point of it is that typing ( requires shift, while [ does not.

https://gitlab.freedesktop.org/xkeyboard-config/xkeyboard-co...

Now, I really could've used something like this on macOS…

Karabiner to the rescue https://genesy.github.io/karabiner-complex-rules-generator/#...

As everybody knows ...

I use(d) arch btw

Ironically, on ISO keyboards, [] need an ALT so even more pain

The Maple syntax may superficially seem easier but actually leads to more problems in practice. The point of the [ ] is that argument of a function is logically distinct from algebraically grouping terms in an equation. Also, Mathematica is a camel case language since underscore is for pattern recognition, hence the capitalization of function names. Personally, I’ve found every little Mathematica design feature to be incredibly well thought out, logical, and consistently implemented over the whole of the language.

I've been at a few universities and labs as a postdoc, and a Mathematica license always came either as part of the University or the department. It might not be relevant in some disciplines, but generally I assume it must be used a lot to warrant such broad licensing (it is a tool I use daily as a theoretical physicist).

You're right -- the theoretical particle physicists at my faculty were using Mathematica very heavily when I was still in academia and maintained a dedicated compute cluster for it.

They really did not appreciate the debugging experience, but maybe that's improved in 15 years. :)

I definitely get the impression that Wolfram builds his tools primarily for himself, and is happy to let other people play with them because that way he gets money to pay for them.

That is not the impression, that is exactly why, And actually that is their strength. Back in the days the whole Apple was there to make software for Jobs and look how awesome that turned out. Wolfram is trying to complete tue work of Leibniz and create a universal calculus. A unifying language for symbolic computation, which is amazing.

I’d like to use it sporadically, but they charge a lot for people in academia, and for my use case it’s simply not worth it.

I’m using xcas now, it’s working pretty well for my humble needs.

It's also free on raspberry pi

While I'm not sure the particular price point is the biggest problem here, the student license pricing doesn't feel seem that great either. The language is hard enough to learn, and most students won't have time to figure out if they want to buy it with a 15-day trial. They'd probably need half a semester at the very least, unless it's a required part of the curriculum. In the rare case where a student is already familiar enough to know they want it, then four years of $75/year is $300... at that point they may as well just pay $390 for a perpetual personal license, so they can at least keep opening their files in the future.

That said, the parent was talking about it being expensive for use in industry. Personal and student licenses aren't relevant there.

Ok fuck it. Bought the year!

My thought is that licenses were similarly cheap for historical programming tools like Turbo Pascal and Visual Basic. My dad got me Turbo Pascal for my birthday, for $39, after reading about it in the Wall Street Journal.

But it seems like the proprietary languages have all withered, regardless of price. Even $195 for Mathematica is an obvious concession to this trend. I don't ever remember it being that cheap.

I could write an essay on the benefits of free tooling, but enough has already been written. I'll spare you the slop. ;-)

Didn’t they also bundle something for free with Raspberry Pi OS?

As discussed on another thread, the outcome is poorly tools glued together, due to lack of roadmap and polish that commercial software usually supports, instead of volunteers coming and going, only caring for their little ich.

I’m not sure about that. I used to use LabView and its various libraries often. The whole thing felt scattered and ossified. I’d take a python standard library any day.

Yet most EE engineers rather use a graphical tool like LabView or Simulink.

Not everyone is keen doing scripting from command line with vi.

Maybe it’s different for those actually working in the profession and n=1 but in my (many) years of studying EE I never used these tools even once.

I once interned at a lab that used a piece of surely overpriced hardware that integrated with Simulink. You would make a Simulink model, and you’d click something and the computer would (IIRC) compile it to C and upload it to the hardware. On the bright side, you didn’t waste time bikeshedding about how to structure things. On the other hand, actually implementing any sort of nontrivial logic was incredibly unpleasant.

Hahaha yep, so much clicking after One day my finger was actually sore.

No not really, depending on the application Cpp or python has been the language of choice in the lab. Labview was used because it was seen as easy to make UIs for operators in production facilities, but even that was a regrettable decision. We ended up rewriting LV business logic in c# and importing it as a lib into a LV front end.

Commercial software polish is lipstick on a pig. A pig that will never be anything else and will eventually die as a pig.

Ugly os software at least has potential to grow internally. Long lived commercial software is a totting carcass with fresh coat of paint every now and then.

Yet, the Year of XYZ software seldom comes, the usual cheering of tools like Blender, often forgets its origin as commercial product and existing userbase.

Someone has to pay the bills for development effort, and when it based on volunteer work, it is mostly followers and not innovators.

There's nothing wrong with commercial software being the origin. What's a crime is that it can stay commercial. Source code should enter public domain in a decade at most.

> What's a crime is that it can stay commercial. Source code should enter public domain in a decade at most.

In many cases, people are free to write their own implementation. Your claim "Source code should enter public domain in a decade at most." means that every software vendor shall be obliged after some time to hand out their source code, which is something very strong to ask for.

What is the true crime are the laws that in some cases make such an own implementation illegal (software patents, probitions of reverse-engineering, ...).

> every software vendor shall be obliged after some time to hand out their source code,

Obviously. Since software is as much vital to the modern world as water, making people who deal with it disclose implementation details is a very small ask.

Access to the market is not a right but a privilege. If you want to sell things we can demand things of you.

You must design your own hardware too, since you can't get the blueprints of commercial products.

Fortunaley hardware designs are routinely reverse engineered and cloned. Imagine the world where industrial designs were as hard to reverse engineer as clone in practice as software. Global GDP would be 10% of what it is. Largest economies of the world owe lion share of their development to cloned designs.

I think commerce between individuals is a right.

Infringing on that should be justified in terms of protecting the rights of those involved, such as ensuring the quality of goods, enforcement of reasonable contract terms and such. We are involved in the process as participants in the market, and that’s the basis of any legitimacy we have to impose any rules in the market. That includes an obligation to fair treatment of other participants.

If someone writes notes, procedures, a diary, software etc for their own use they are under no obligation to publish it, ever. That’s basic privacy protection. Whether an executable was written from scratch in an assembler or is compiled from high level source code isn’t anyone else’s business. It should meet quality standards for commercial transactions and that’s it. There’s no more obligation to publish source than there is to publish design documents, early versions, or unpublished material. That would be an overreaching invasion of privacy.

So restrict the obligation to companies.

On what justification? You just want to take their stuff, because?

People shouldn’t lose their rights to what they own, just because they do so through a company.

I do think reasonable taxation and regulation is justifiable but on the understanding that it is an imposition. There is a give and take when it comes to rights and obligations, but this seems like overreach.

> Obviously. Since software is as much vital to the modern world as water, making people who deal with it disclose implementation details is a very small ask.

The analogy would be ever-so-slightly more accurate if you said "software is as much vital to the modern world as beverages".

It would also be more accurate if all water was free.

Neither of which is the case.

I see, you think actively preventing companies and individuals from interacting freely is the default. And it's a privilege allowed to?

Well I wonder who it is you think has the right to deny others the freedom to cooperate economically by default. Then, allow "privileges" so people can work together.

Aside from that moral upside-down world, what you are describing is a steep limit on copyrights, with forced source, i.e. trade secret, reveals.

So you are removing the huge incentives that copyright creates. If software were always trivial to build, or cost very little to build, that would not be problem. In real life, that would devastate software work, and we would all be poorer for it. Companies, individual software developers, and users.

Not everyone buys into FOSS religion, especially when there are bills to pay, and too many people feeling entitled to leech on work of others and being paid themselves, or companies for that matter.

Worse than lipstick on a pig is lipstick all the way down, with no pork, like the user interfaces coming out of Apple.

To be fair, "sundry tools poorly glued together" describes CAS and symbolic computation software in general, including Maple or Mathematica. It's surprisingly difficult to put a proper formal foundation (guaranteeing the absence of "wrong" or even outright meaningless results) even on very basic symbolic manipulations.

> If this was open sourced, it had the potential to severely change the software/IT industry.

As an engineering undergrad I had a similar feeling about Matlab & Mathematica.

Matlab especially had 'tool boxes' that you bought as add-ons to do specific engineering calcs and simulations and it was great, but I almost always found myself recreating things in python just because it felt slightly more sane to handle data and glue code.

Pandas and Matplotlib and SciPy all used via an ipython notebook were almost a replacement.

Actually, Wolfram Language is based on Tiny Caml. M-expression based though. Lisps are neat and cosy too.

There is of course a FOSS rewrite https://mathics.org

Interesting. I have always felt I am missing out on not using tools like Mathematica or MatLab. I see some people doing everything using MatLab, including building GUI and DL models, which I found surprising for a single software suite, and - nowadays - one that is quite affordable (at least the home edition).

Mathematica seems a little pricey but maybe it would motivate me to learn more math.

I would love to read what non-mathematicians use MatLab, Mathematica, and Maple for.

I'm a non-mathematician and I used it for lots of novel stuff - GIS, visualisations of all kinds, machine learning. The Wolfram Community staff picks is a great introduction into the varied things you can do: https://community.wolfram.com/content?curTag=staff%20picks

I used mathematica for real last time in SGI days and loved it. I know probably a ton has changed since, but I do have to ask those that use it today if you'd still use it for non math-heavy (and even so) tasks if you have access to the wonderful world of python and jupyter / polars, R, and similar?

Mathematica is awesome for weird, one-off tasks in fields that I'm unfamiliar with, since the documentation is excellent, and the functionality is really broad (so I don't need to figure out how to install a specialty program for every one-off task). But for fields that I'm experienced with or tasks that I'm planning on running frequently, I'll usually just use Python, since most of the Python libraries have more functionality and run quicker than Mathematica.

(Mathematica is of course much better than Python at symbolic math, but this isn't what you are asking about)

To be fair it was used a lot during my physics studies. I opted to use it afterwards for integrals and derivations, very powerful.

Yeah, I was one of those schmucks that used sympy / python instead of mathematica in my physics coursework. Policy was "mathematica is recommended and supported, but you can bring your own tools if you want to and can make them work."

In retrospect, doing the work in mathematica would have probably stretched my brain more (in a good way!) since it provides a different and powerful way of solving problems vs other languages...maybe I'll have to revisit it. Perhaps even try advent of code with it?

While python did get the job done, it feels like the ceiling (especially for power users) is so much higher in mathematica.

MatLab was taught and used extensively at my university, and has many strong sides and a fantastic standard library. We used it mainly for physics and robotics calculations. The licenses are (were?) prohibitively expensive outside of academia though. Hard to compete with free Python + NumPy and a larger talent pool.

Matlab and Mathematica are VERY different.

Matlab and Python are in the same ballpark. Easy syntax and large standard library. Matlab provides a lot more dedicated libraries for niche areas but the overall experience feels the same.

Mathematica doesn't really have a standard counterpart. Jupyter notebooks try to capture the experience but the support for symbolic expressions makes the Mathematica experience very different.

Matlab is fundamentally not a great language. It's not great at interacting with data that isn't already blocks of numbers, it's terrible for UI design, and even for matrix manipulation numpy is beating it nowadays. The main reason to use it is if you really need something in one of the toolboxes that doesn't have an open-source equivalent, or for Simulink. As a general data processing language it's far from the best default.

I've had it installed on my laptop for over two decades, and I use it maybe once a year for actual work. Every time, it feels like cracking a walnut with a 500-ton press.

Mathematica is way, way under appreciated in industry, and even in the sciences.

> Nowadays I'd probably just ask Claude to figure it out for me

Incidentally, Mathematica + LLMs make a great combination. If you take what is pretty much the biggest mathematical routine library in the world and combine it with interactive visualization tools, and then use an LLM to accelerate things, it becomes an incredible tool. Almost ridiculously powerful for trying things out, teaching, visualizing things, etc.

(I've been using Mathematica since 1992 or so, so I'm familiar with the language, but it's still so much faster to just tell Claude to visualize this or that)

Yes, I’m sure! I had mostly stopped using WL by the time ChatGPT came out and so my only experience of their integration is a lot of hallucinated syntax by GPT3. I’ve not been tempted to upgrade to the newer, more LLM-integrated versions yet, particularly because I’ve been trying to cut down on expenses and spending more money on LLM api calls versus just using my existing Anthropic subscription for Claude code isn’t that appealing. I’ve also been going through a bit of a crisis of a lack of creativity and inventiveness and it’s hard to decide to spend money on an update when it doesn’t seem like I have any good ideas of what to do with it after I’ve spent the cash!

I bet that if they can integrate LLMs _really_ well (I’m not sure the chat driven notebook thing is necessarily the way) it’ll be a massive upgrade.

The tech world needs more Stephan Wolframs.

Yes this is all a bit quirky and nuanced but when you get into it these things are really good. It’s refreshing to see some really smart folks just focused on doing great things without blinding from VCs and MBAs pushing another hacky quick way to make a buck and cash out.

I agree. Wolfram gets so much crap ok this site but I think he’s produced so much interesting science and tech that has built up over decades, that’s very admirable.

Maybe with the power of a supercomputer, Mathematica can finally launch in less than 30s. I have no idea how a software that still does essentially the same thing as it did in 1988 can be that sluggish.

Yet there is hardly any computing system that can replicate Mathematica tooling capabilities.

One would expect 37 years would be enough to create such alternative.

Jupiter notebooks aren't the same.

Honest question: I think Maple is really a competitor, even if less known.

Prob I am a very small minority here, but I used Maple a lot in the past and I liked it a lot. More standard syntax than mathematica, and overall much much easier to debug. In particular, as far as I remember, maple allowed for more transparency in the mathematical methods used in some computation than mathematica, as I could just see what it was doing exactly when an unexpected result came or I wanted to understand some method. When you work a lot with integrals, limits and special functions there is always a shit ton of assumptions and cases all over the place, and if something is missed somewhere the computation can just get wrong. There were some few times I would actually get wrong results in mathematica. When the language is more like a black box, it is hard to know what happened.

That is also my feeling, although it has been ages since I last worked with Maple. But some colleagues of mine at other uni do use it instead of Mathematica. Thanks for the detailed reply!

Sage Math? Though I admit, unlike homogeneous Mathematica, it's just a Python glue on multiple smaller projects of different quality and poorly integrated. I wish there was something more like the Wolfram software but there isn't.

I quite like Sage. Python is a much better language than Wolfram (yes, he named it after himself...). In Wolfram, there is no real scoping (even different notebooks share all variables, Module[] is incredibly clumsy), no real control flow (If[] is just a function), and no real error handling. When Wolfram encounters an exception, it just prints a red message and keeps chugging along with the output of the error'd function being replaced by a symbolic expression. This usually leads to pages and pages of gibberish and/or crashes the kernel (which for some reason is quite difficult to interrupt or restart). Together with the notebook format and the laughable debugger, this makes finding errors extremely frustrating.

The notebooks are also difficult to version control (unreadable diffs for minor changes), and unit testing is clearly just an afterthought. Also the GUI performance is bad. Put more than a hand full of plots on a page, and everything slows to a crawl. What keeps me coming back is the comprehensive function library, and the formula inputs. I find it quite difficult to spot mistakes in mathematical expressions written in Python syntax.

Fully agreed. I have never seen a programming language which is so badly designed as Wolfram. I really wish there was another way to access all of Mathematica's functionality with a more sane interface.

> Python is a much better language than Wolfram

Different languages are better at different things, so it rarely makes much sense to say that one language is better than another in general. Python is definitely much better than Mathematica for "typical" imperative programming tasks (web servers, CLI programs, CRUD apps, etc.), but Mathematica is much better at data processing, symbolic manipulation, drawing plots, and other similar tasks.

> there is no real scoping (even different notebooks share all variables, Module[] is incredibly clumsy)

Scoping is indeed an absolute mess, and the thing that I personally find the most irritating about the language.

> no real control flow (If[] is just a function)

You're meant to program Mathematica by using patterns and operating on lists as a whole, so you should rarely need to use branching/control flow/If[]. It's a very different style of programming that takes quite a while to get used to, but it works really well for some tasks.

> no real error handling

For input validation, you should use the pattern features to make it impossible to even call the function with invalid input. And for errors in computation, it often makes the most sense to return "Undefined", "Null", "Infinity", or something similar, and then propagate that through the rest of the expression.

> The notebooks are also difficult to version control (unreadable diffs for minor changes)

Mathematica notebooks tend to do slightly better with version control than Jupyter Notebooks, although they're both terrible. You can work around this with Git clean/smudge filters, or you can just use ".wls"/".py" files directly.

On the note of Jupyter notebooks and version control - there was a talk at this year's Pycon Ireland about using a built in cleaner for notebooks when committing the JSON (discard the cell results), and then dropping the whole lot into a CI system utilising remote execution (and Bazel or similar) to run and cache the outputs. Was a talk from CodeThink. No video up yet though. Scenario was reproducible notebooks for processing data from a system under test.

> On the note of Jupyter notebooks and version control - there was a talk at this year's Pycon Ireland about using a built in cleaner for notebooks when committing the JSON (discard the cell results)

Yup, I use a long "jq" command [0] as a Git clean filter for my Jupyter notebooks, and it works really well. I use a similar program [1] for Mathematica notebooks, and it also works really well.

[0]: https://stackoverflow.com/a/74104693

[1]: https://github.com/JP-Ellis/mathematica-notebook-filter

For writing production code, I find good scoping rules non-negotiable. And error handling, monitoring etc has to be well thought out before deploying at scale.

So as great as Mathematica sounds for interactive math and science computations, sounds like a poor tool for building systems that will be deployed and used by many people.

That is a fair assessment. By and large it is used for the former. It is super handy in the exploratory phase of certain kinds of mathematical research.

> So as great as Mathematica sounds for interactive math and science computations, sounds like a poor tool for building systems that will be deployed and used by many people.

Yes, I definitely agree there. Mathematica is definitely great for interactive use, but I'm not really aware of anyone aside from Wolfram himself who tries to deploy it at scale.

How is "If" as a function even a drawback? It is largely seen as something desired, no? I would see that as a huge advantage, which allows for very powerful programming and meta-programming techniques.

One potential issue is that unlike most other languages, it doesn't create a new scope. But almost nothing in Mathematica introduces a new scope, and Python also uses unscoped "if"s, so it's rarely much of a problem in practice.

But with pattern matching, you almost never need to use "If[]" anyways:

  fib[0] := 0
  fib[1] := 1
  fib[n_ /; n < 2] := Undefined
  fib[n_Integer] := fib[n - 1] + fib[n - 2]

  fib[8]
  (* Output: 21 *)

  fib /@ Range[10]
  (* Output: {1, 1, 2, 3, 5, 8, 13, 21, 34, 55} *)

  fib[-1]
  (* Output: Undefined *)

  fib["a string"]
  (* Output: fib["a string"] *)

This is not true. Mathematica has the concept of contexts. You can have each notebook have it's own unique context. Mathematica Packages create their own context too, we are not talking about module's here which are useful for local variable scoping. Packages and contexts lead to the isolation you are looking for. These are things that have been around since the initial Mathematica 1.0 in 1988 (!). https://reference.wolfram.com/language/ref/Context.html

Same about your criticism of error handling and control flow: https://reference.wolfram.com/language/guide/RobustnessAndEr...

Yeah, but that was my point, similar capabilities, not a bunch of tools glued together, poorly integrated, as you point out.

Have you tried Maple? It is also proprietary but it comes from (originally) Waterloo University, and is used in Academia too.

I tried Sage Math. Just the fact that one has to declare all variables before using them, makes it extremely annoying. In Mathematica, I frequently do computations which have a couple of dozen variables. I am not going to write boiler plate for 20 different variables in every notebook.

I've used Sage for years to run the backend (calculations/computations/graphics/prototyping) for a multivariable calculus class I teach. It's not perfect, but as a lightweight, Python-style CAS to do all sorts of "standard" calculations, it's very easy to use!

I think it's an extreme example of not-invented-here syndrome. In many ways that leads to interesting novelties, and in others it leads to not having undo/redo until the 2010s.

In fairness the problem is not easy! JupyterLab only got document-wide undo/redo in 2021 per https://jupyterlab.readthedocs.io/en/3.1.x/getting_started/c... .

Mathmatica has been around since the 80s, Jupyter how long? Far leas.

"Fun" fact: This is partially because Wolfram owned a patent on the "code notebook" invention. Once it expired, Jupyter was created.

This is how I feel about basically all consumer software

So "use our proprietary service to scale our proprietary language" is a great pitch for people who are already all in. Increasing spend among existing customers won't help you get new ones though. And it kind of feels like a prelude to nerfing non-cloud based usage.

Typical example of a extraction/exploitation mentality where innovation would be better. Wolfram is in an amazingly good spot to spin up better "simulation as a service" if they would look at fine-tuning LLMs for compiling natural language (or academic papers) into mathematica semi-autonomously and very reliably. Mathworld is potentially a huge asset for that sort of thing too.

Good point

I love this, too many people keep themselves busy discussing bits and bytes, or Stephen Wolfram's personality.

The reality is that by now we should already be at a level where common programming would be like Wolfram everywhere.

Maybe agents and LLM driven code generation is how we eventually get into the next abstraction level, sadly won't be without casualties with smaller team sizes, when so much can be automated away.

> The reality is that by now we should already be at a level where common programming would be like Wolfram everywhere.

What aspects of the Wolfram language should be everywhere? The easy access to lots of datasets? The easy access to lots of mathematical functions? CAS in general?

All of those, plus having an interactive graphical environment, debugging experience and optimisation when needed.

Basically the ideas of Smalltalk and Lisp Machine variations, that are still only partially available in modern IDEs, and proudly ignored by the "vt100 rules and vi first" minded devs.

Yes!

Reminds me of the “Stop writing Dead Programs” talk. https://news.ycombinator.com/item?id=33270235

Hi Stephen, is that you?

Jokes and sales pitches aside. We kinda have that already, we platforms that allow us to run the same code on, x86, arm, wasm… and so on. It’s just there is no consensus on which platform to use. Nor should there be since that would slow progress of new and better ways to program.

We will never have one language to span graphics, full stack, embedded, scientific, high performance, etc without excessive bloat.

Many VCs today would vouch that would be agentic AI.

No. I work in VC backed startups. Externally they might say that for investors, but talk to a software engineer. Agentic IA is not working, it will regurgitate code from examples online, but if you get it to do something complex or slightly novel, it falls flat on its face.

Maybe it will someday be good enough, but not today, and probably not for at least 5 years.

Meanwhile there are people delivering solutions in iPaaS tools, which is quite far from the traditional programming that gets discussed on HN.

Some of those tools aren't yet fully there, but also aren't completely dumb, they get more done in a day, than trying to do the same workflows with classical programming.

Workato, Boomi, Powerapps, Opal,...

How is that relevant to this thread?

Those are high level programming tools, nowadays backed by AI, being adopted the big corps, some are sponsored with VC money, which you are sceptical about.

Well good luck to them. I don’t want to be stuck without a job because the bubble burst along with all the capital my job depends on.

I really like this way of introducing a new feature/service. Straight to the point, explains what it does, which problem it solves, gives practical examples and walks the reader through them. So many times when I read about a new feature/service I am left with more questions than I had started with, but this was great!

One of the more interesting things about WL is that Stephen Wolfram is really a genuine daily user of the software his company makes and he's the final say on what ships and in what form. They used to livestream his meetings reviewing potential new features on Youtube, an interesting watch. It didn't make me want to work there but I did feel like he cared very much. Quite Jobsian, dare I say.

> Quite Jobsian, dare I say.

Good for product: not so good for people.

I am told that he gave a great deal of agency to people he trusted, though.

In my career, I ran into two [brilliant] individuals that had, at one time, worked with Jobs.

They both hated him.

Huh. So Stephen finally discovered cloud computing (yes, I know about the hosted notebooks).

I played around with RemoteKernel some time ago (https://taoofmac.com/space/blog/2016/08/10/0830) but this is “better”, although I wish they’d make it hostable in your own cloud provider like materials simulation software and other things we see running in HPC clusters. (I also ran Mathematica in a 512GB/128core VM once for kicks, but it’s just not cost-effective).

Just before I stopped using Mathematica they came out with that headless kernel, and I had wondered if you could spin it up on a Kubernetes cluster or something.

I do notice that they have an "Application Server" for Kubernetes, which is pretty curious: https://github.com/WolframResearch/WAS-Kubernetes (though not updated in over a year)

> I wish they’d make it hostable in your own cloud provider

Wolfram Compute Services is based on the RemoteBatchSubmit featureset, which also supports AWS Batch (https://reference.wolfram.com/language/ref/batchcomputationp...) and Azure Batch (https://reference.wolfram.com/language/ref/batchcomputationp...) backends.

[Disclosure: I am the main developer of RemoteBatchSubmit.]

With all the integrated standard functions Mathematica is such an incredible tool. We really need an open source version of it. Even if we implement only 10% of the features it would be already incredible useful.

I started working on an implementation in Rust called Woxi (https://github.com/ad-si/Woxi) and I hope to find some contributors, as it is such a gargantuan task!

Here's a 10% clone of Mathematica - https://mathics.org/

Maxima is the clone. I remember using it 20 years ago. very nice.

Maxima, despite its shortcomings, is a great tool for symbolic computation.

It's more like both Mathematica and Maxima are clones of the original MACSYMA.

Maxima is based on the DOE (Department of Energy) fork of MACSYMA.

Maxima is a similar CAS, but it's not a clone of Mathematica like Mathics aims to be.

This reminded me of one of the greatest posts in the history of the internet, and I am devastated to report that it no longer exists: https://news.ycombinator.com/item?id=6516114

imo you should implement double pass for the evaluator. using pest tokens in the functions etc makes it gard to implement also makes it slow and not optimizable

Yes agreed! I already started implementing it for some subset. Hope to get it finished soon.

I've writing one myself in Go the past few months just to learn the way CAS internals work. Got about 300 symbols working to varying degree. Complex numbers almost working. Basic rule system. Various algebra simplifications. Still trying to figure out how to do Factor as it is pretty math intensive. No graphical output, just to the terminal for now. Using Claude models also to learn AI programming and help with so of the math concepts.

I've been using Mathematica since 1992 or so and at least for the last 20 years I've been wondering why they didn't provide cloud computing services, so that I can simply evaluate stuff remotely on a larger computer. So, I guess I should say "FINALLY!" :-)

Do quant traders use Mathematica? I would guess this would be a great use case for a tool that lots of people love. Pretty language, huge boatload of built-in tools, high powered mathematics, great visualization capabilities. Quant firms should be able to live with the price tag. I assume they have a compiler that can produce fast executables for HFT.

They do

Instead of offloading batch computations to a proprietary cloud, it’s better to actually optimize the incredibly slow and unstable computational kernel.

In any case that’s not the happy path, Mathematica gets stuck in symbolic computations for ages. My FFT-based research in Mathematica slowed to a crawl, tens of minutes of waiting, even with 90% of the code compiled to binary. MATLAB finishes this task in milliseconds.

the beauty of pure functions :)

I'm slightly terrified of the combination of Wolfram and a LLM.

I’m not worried, wolfram is already a huge language model, adding a large one won’t make it any better/worse :-)

What I know about Wolfram is pretty old, so maybe I was unaware. But I do remember a time before LLMs and even 15 years ago, Wolfram was very impressive.

Reply on news.ycombinator.com