TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Starting to Demo the Wolfram Language [video]

190 点作者 samolang超过 11 年前

23 条评论

primitivesuave超过 11 年前
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&#x27;s an incredible and fascinating tool when you&#x27;re given these toy models to put into it, but I don&#x27;t really see a Wolfram Language revolution happening anytime soon.<p>One aspect I think Mathematica will excel in is code generation. Using symbolic constructs in a very high level language to generate constructs of a low-level language is much easier to do in Mathematica than basically any other language.<p>On a side note - Linus Torvalds thought naming his operating system &quot;Linux&quot; was too arrogant. I think it&#x27;s pretty bad marketing to name your programming language after yourself, but then again, the same social conventions that apply to us regular human beings certainly don&#x27;t apply to Stephen Wolfram.
评论 #7294968 未加载
评论 #7294387 未加载
评论 #7294501 未加载
评论 #7296202 未加载
nswanberg超过 11 年前
This has been a very long time in coming. Here&#x27;s Wolfram in a 1993 interview:<p>&quot;One of the things that I consider an exciting direction is to what extent we can expand the use of the language itself, independent of the application side of Mathematica. We&#x27;ve considered making a thing that will probably be called M, that is essentially Mathematica without the mathematics.&quot;<p><a href="http://www.stephenwolfram.com/media/stephen-wolfram-multiparadigm-man/" rel="nofollow">http:&#x2F;&#x2F;www.stephenwolfram.com&#x2F;media&#x2F;stephen-wolfram-multipar...</a><p>What&#x27;s funny is that it&#x27;s only the &quot;without the mathematics&quot; part that hasn&#x27;t necessarily come true, except possibly in terms of marketing. This little Raspberry Pi below my desk has a &quot;Wolfram Language&quot; distribution on it, which as far as I can tell is exactly the same as Mathematica.
评论 #7296676 未加载
nl超过 11 年前
I&#x27;m working on something akin to this, although my version is somewhat more ambitious[1].<p>So far I have a &quot;Knowledge Engine&quot; 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).<p>I&#x27;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.<p>My version is designed to be completely private, open source and self-hostable, including all the data[1], so I can&#x27;t rely on AppEngine. Instead, I have SymPy running inside ZeroVm + LXC (Docker) as a sandbox. I&#x27;m currently working out how to get that to communicate with the front end (which is non-trivial, because ZeroVm is pretty much undocumented).<p>Obviously SymPy isn&#x27;t as &quot;Natural Language Like&quot; as the WolframAlpha language, but arguably Python is actually more powerful.<p>After that I&#x27;ll do OpenStreetMap integration (ie, self hosted OSM), which will give it better geographic understanding that it has.<p>[1] Every time I mention this I need to include a standard disclaimer that I do realize this is completely and utterly crazy.<p>[2] <a href="http://sympy.org/en/index.html" rel="nofollow">http:&#x2F;&#x2F;sympy.org&#x2F;en&#x2F;index.html</a><p>[3] <a href="http://live.sympy.org/" rel="nofollow">http:&#x2F;&#x2F;live.sympy.org&#x2F;</a>
评论 #7295753 未加载
hardwaresofton超过 11 年前
This seems like they have essentially codified wolfram alpha&#x27;s many APIs and created the biggest, baddest standard library any language has ever seen, and slapped on some functions to access everything with. It seems like the wolfram language essentially sits on abstraction level max-1 which is pretty amazing.<p>I wonder how things break down when you don&#x27;t want to do something that can be perfectly described in existing functions (ex. you need to perform some novel computation on an intermediate result of some function). Though I&#x27;m sure people much smarter than myself have made sure the code is modular enough to allow you to get your hands dirty (or they written the code such that if you have to get your hands dirty, you&#x27;re doing it wrong)
评论 #7295764 未加载
评论 #7296351 未加载
DanielBMarkham超过 11 年前
First, as random internet commentator guy, I want to thank Stephen for his contribution here. This and NKS are incredible.<p>Second, I keep seeing videos and articles about Wolfram Alpha and am always left scratching my head kinda going &quot;WTF?&quot;<p>So it&#x27;s a knowledge-based purely symbolic programming language? Can it play Flappy Bird? Power the next cool startup? Give me an answer to a question I might have today? I&#x27;ve been to alpha a couple of times. Never seemed to be able to get the questions right to get the answers I wanted.<p>I don&#x27;t mean that to be negative. My point is that any audience is going to have a thousand different needs and viewpoints, most of which will probably seem trivial to Stephen Wolfram. Most new technologies get past this hurdle by open-sourcing. That way there&#x27;s a thousand experiments, and over time a few of them are bound to give all of us little nerds a gleam in our eye one way or the other.<p>The other way of doing this, where you have a single point of creation describing to the world how cool it is? It is limited to things that have a single terrific property that everybody can agree on.<p>I&#x27;m not sure that is the case here. Still -- awesome stuff. Can&#x27;t wait to watch this evolve.
评论 #7294492 未加载
评论 #7294203 未加载
评论 #7294430 未加载
评论 #7296748 未加载
tunesmith超过 11 年前
As someone that wasn&#x27;t a math major, I regularly get excited about Mathematica but then have trouble figuring out why I would actually want to use it. If someone were to give me an assignment that <i>they</i> knew could be accomplished with Mathematica, then that would be easier to start, but I feel like it takes a while to make that translation between having a vague desire to figure something out, and then realize that Mathematica could do it for me quickly. Like, I wouldn&#x27;t have thought to plot out last year&#x27;s trip to Europe using it. So for new users, there&#x27;s something of a &quot;well sure it&#x27;s easy AFTER you know how to do it&quot; gap.<p>It kind of reminds me of when search engines first came out - I regularly witnessed a gap between people having fun typing in some example queries, and realizing that they could type in <i>anything</i> search-engine-ish and get back information they actually didn&#x27;t know ahead of time; information they had conditioned themselves not to ask out of not wanting to schedule an afternoon trip to the library.
评论 #7298539 未加载
gklitt超过 11 年前
Wow, nice demo. It is interesting to see Wolfram explain his vision for an &quot;ultimate fusion language&quot; -- a powerful multi-paradigm language, a huge standard library, data sources, and deployment all rolled into one. This monolithic approach seems to be the opposite of the Unix philosophy, and I wonder how that will play out.<p>There are clearly benefits to be gained from smooth integration, and the demo is impressive considering it only uses stuff built into the language. But there&#x27;s also something to be said for letting different teams of people focus on building smaller and more focused tools which can be combined in a variety of ways. I worry that this Wolfram language will be limited in its usefulness for real applications because it simply tries to bite off too much.
评论 #7296715 未加载
评论 #7307625 未加载
评论 #7294777 未加载
sadfaceunread超过 11 年前
The best parts are the natural language inputs and inherent connectedness with data. These are going to eventually become expectations that you&#x27;ll have interacting with computer systems in general.<p>Having a computer interface that lets you say, find out who in my office is sending the most email, in a one liner is the next logical step. Including and understanding your company &#x2F; personal data in the way Wolfram Alpha understands its curated data is the missing link.<p>I&#x27;d be really interested in integration with enterprise resource planning software, and customer relational management software for big corps. For web firms an interface to salesforce.com, an interface to your web analytics data, your support ticket dashboard, your web marketing data. I imagine being able to simply ask questions like how does the weather in my customers location impact my conversion rate? Or other poignant questions mixing internal and external data seamlessly.
dkural超过 11 年前
This is really fantastic of course, but the video doesn&#x27;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&#x27;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.
prezjordan超过 11 年前
This seems super exciting. I&#x27;m not sure if you could build large-scale applications with it, but I cannot wait to play around with it.<p>Side note: The use of &quot;I&quot; and &quot;My&quot; really irks me. I imagine many, many people made this happen and contributed to these design goals. I guess this shouldn&#x27;t take away from how cool this technology is, but it definitely leaves a bad taste in my mouth.
评论 #7294755 未加载
评论 #7294457 未加载
beachstartup超过 11 年前
one use case could be a business analytics tool. &#x27;business intelligence&#x27;.<p>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.<p>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.<p>i think in the future, managers&#x2F;executives who will separate themselves from the MBA herd are the ones who will be able to effectively use tools such as this.
archagon超过 11 年前
This demo is pretty amazing and will take me a while to digest.<p>A few questions that I&#x27;m really interested in hearing the answer to:<p>1. Will the language work without an internet connection?<p>2. Will it be free to use?<p>3. What happens if&#x2F;when Wolfram the company goes out of business, or Wolfram himself retires?<p>4. Are all the disparate APIs and sources that the language uses determined by fiat, or can they be customized? What happens if&#x2F;when the APIs&#x2F;sources stop working?<p>5. Given that the language can do things like query your Facebook graph, it seems like there&#x27;s a lot of API upkeep that the language requires. Is this done by Wolfram? How does this work when I&#x27;m running locally (if this is even possible)?
phantom_oracle超过 11 年前
It was once said that Stephen Wolfram builds such powerful tools that even the great man himself doesn&#x27;t know how to use or explain what these tools are :P<p>On a more serious note, think of this language as something highly-sophisticated tech and other-industry companies might use.<p>You won&#x27;t be building a farting-noise-type app with this language, unless it is a bowel-detecting-smell-sniffing-doctor-replacing-treatment-providing analysis app that uses sensors to compute the above scenario.<p>If Wolfram wants the language to be used in knowledge and computation fields, then maybe a couple of demos on how to use it for something like stock-trading, or tweet-timing-reach or weather-analysis might get this into the hands of people that use computation in some form with rather &#x27;primitive&#x27; tools when compared to the Wolfram language (which basically is the equivalent of a huge-ass library and works like a huge-ass framework of sorts).<p>I personally would probably like to try or see the language being used in optimizing an industry like logistics or travel or anything to do with maximizing time-efficiency.
iandanforth超过 11 年前
So while I can&#x27;t see myself using this to write applications I can certainly see it replacing &#x2F; augmenting my current iPython notebooks. I can do <i>a lot</i> with pandas and plotly for data exploration and visualization, but I can&#x27;t do many of the things he demo&#x27;ed nearly as easily.
diakritikal超过 11 年前
Ego Force One has taken off again :(
评论 #7294913 未加载
评论 #7296135 未加载
garysweaver超过 11 年前
This might be a good place to demo the language, assuming it is meant to be general purpose:<p><a href="http://rosettacode.org/wiki/Category:Programming_Languages" rel="nofollow">http:&#x2F;&#x2F;rosettacode.org&#x2F;wiki&#x2F;Category:Programming_Languages</a>
lmg643超过 11 年前
Any ideas of how the Wolfram Language can be re-used in other commercial applications? Seems pretty powerful capability, in theory you could build a ton of services on the back of this, much like google, only with some new interesting hooks.
评论 #7294539 未加载
DonGateley超过 11 年前
Wow! That is a helluva pitch. My fear is that it _is_ the mind of Stephen Wolfram and unless you have that the whole thing will be one big mystery.
评论 #7296691 未加载
DonGateley超过 11 年前
I would sure like to see the code for its repl. If it&#x27;s all he says, that loop should be uniquely tiny.
评论 #7307788 未加载
nicklovescode超过 11 年前
Well all my programming tools feel obsolete.
curious222超过 11 年前
How to run it? Is there any interpreter?
mmanfrin超过 11 年前
This is really, really cool.
pjmlp超过 11 年前
Lisp Machines are back! :)