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

Wolfram Alpha doesn't "get it"

blog.agiletortoise.com

Post reply on HN
Print view View original

101–110 of 111 posts

Re: Wolfram Alpha doesn't "get it"

#101

Post by mcantor »

nyellin wrote: post #46
kragen wrote: post #38

Earlier quoted context omitted.

There's a censored comment in response here from the original author of the post. Turn on showdead.

Is there any way to see the justification for killing a comment? I trust HN's moderators, but it makes me uneasy to realize that there is a whole collection of censored comments that I didn't see until now, because I had showdead disabled.

I'm curious about some threads that get killed, too--afaik it can happen due to mod intervention or when the community hits "flag" enough times, but I always wonder which one it was.

Re: Wolfram Alpha doesn't "get it"

#102

Post by ugh »

mc32 wrote: post #69
kbutler wrote: post #53

The problem with the Wolfram Alpha terms of service is that they forbid natural, common uses of the web site. "The Wolfram|Alpha service may be used only by a human being using a conventional web browser to manually enter queries one at a time." Think for a minute what that forbids. Generating links programmatically, yes. Tweeting links? Yes. Emailing a URL? yes. Bookmarks? Oh, yeah... Enforceability of the ToS aside…

>The problem with the Wolfram Alpha terms of service is that they forbid natural, common uses of the web site. It seems that that only applies to general circumstance usage. That is, it appears, once can enter into arrangements with W|A to allow non-human usage of website. To prove, just search for "calories in apples", for example in Bing or Google search.

I have to enter into an agreement with W|A if I want to bookmark a page?

Re: Wolfram Alpha doesn't "get it"

#103

Post by joelhaus »

scott_s wrote: post #2

Please submit full URLs only. We like to know where we are going.

[off-topic] Assuming it was a shortened link, Twitter's new interface solved this issue by dynamically generating title="" tags that display the full URL on:hover.

Anyone aware of a good solution for this? Shortener API's + An open source script?

Re: Wolfram Alpha doesn't "get it"

#104

Post by limmeau »

abcd_f wrote: post #96
StrawberryFrog wrote: post #92

Earlier quoted context omitted.

The problem with the Wolfram Alpha terms of service is that they forbid natural, common uses of the web site. Exactly. For instance, http://www.wolframalpha.com/input/?i=does+wolfram+alpha+get+...

And the answer is Sometimes - http://i53.tinypic.com/2vkhrt2.gif That's a good one, W|A :)

I hope you typed in the URL manually.

Re: Wolfram Alpha doesn't "get it"

#105

Post by Stormbringer »

The problem is twofold:

(1) semantics (fun for the whole family!) what does deep linking mean in this context? We all say deep linking is bad, but to some the pre-populated query isn't deep linking, but to others it is.

(2) one of the (entirely sensible) things the Wolfram Alpha guys seem to be saying is "please don't hook bots up to us". Now no doubt some will bluster and say that if it is there in public they have a right to consume all the bandwidth and resources - after all if they didn't want people to do that they should just [insert complex but fragile technological solution that would quickly fail or be worked around]. As it happens, what this guy was doing falls into that category of automated client.

So on the basis of either one of those (if you squint real hard) I can see Wolfram Alpha's point of view.

Re: Wolfram Alpha doesn't "get it"

#106

Post by _b8r0 »

It seems to me that agile tortoise doesn't "get it". Regardless of Wolfram Alpha's licence, it's Wolfram's decision as to what you can and can't do. That was made clear in the licence.

The bottom line is that Agile Tortoise didn't study the licence hard enough, if he had he would've seen that Wolfram Alpha wasn't the right fit for his tool. If he'd have looked further he'd have probably seen that http://duckduckgo.com/ has an API that he may well have been able to use to get the same results.

Instead he may now have to modify his software and probably resubmit it to Apple for review.

Measure twice, cut once. This applies as much to software licensing as carpentry.

Re: Wolfram Alpha doesn't "get it"

#107

Post by mcherm »

agiletortoise wrote: post #27
kenjackson wrote: post #16

Earlier quoted context omitted.

but in this case agiletortoise didn't gain any benefit from the website -- he simply wrote a program that directed the user's browser to go there. I don't get that. Doesn't agiletortoise get benefit when their users get benefit?

Yes, I got a benefit. There were users that like the Wolfram links and will be sorry they are gone. As I said in my post, I'm fine with it. I think Wolfram's Terms are OK. I think it's a bit tenuous only in that if they were really serious about it, they could enforce that requirement in their code. They are not doing that, because it's to their benefit to allow some linking -- they are just reserving the right to se…

I think there is a very plausible argument to be made that it was your CUSTOMERS who benefited, not you.

In other words, I think that if Wolfram chose to sue you for thousands of dollars plus court costs because you violated your contract with them that you would be able to say, in defense, "I never agreed to anything." Since Wolfram delivered the goods (some answers, in this case) to the USERS of your application, not to you, I think this might be a valid defense.

Of course, the strictures of polite behavior extend further than the iron force of the law. Polite behavior suggests that when they ask you not to link to them you shouldn't... which is exactly what you did.

Re: Wolfram Alpha doesn't "get it"

#108

Post by Someone »

jcarreiro wrote: post #40
raganwald wrote: post #10

Earlier quoted context omitted.

As I read this, he's not making links to their web pages. He's deep linking to their content and displaying it within what appears to be a custom browser. While there is plenty of debate about the propriety of deep linking and whether sites have the right to bar deep linking, displaying content within a custom browser is analogous to 'framing' content without permission, and framing content a no-no. Creating deep lin…

What is the difference between "linking" and "deep linking"? Whatever else W|A is, it is a web server and it responds to my requests by sending me web pages. Why does Wolfram get to decide for me which web browser I should use to view those pages?

It is similar to the difference between "citing" and "copying"/"plagiarizing": a gradual thing where the distinction is clear at extreme ends of the scale, but muddy in the middle.

Re: Wolfram Alpha doesn't "get it"

#109

Post by pbhjpbhj »

kragen wrote: post #91
pbhjpbhj wrote: post #56

Earlier quoted context omitted.

I like your example but I don't think that a web page is like a person in public space. As you say their is no tort generally in approaching someone in the street but it is possible that a legal barrier has been erected and so an uninformed observer can't assume that no infringing/illegal activity is being undertaken (think exclusion orders, stalker laws, anti-social behaviour orders and the like; I think these are r…

> Being able to view the site by making a temporary copy in your computer's cache is not clearly non-infringing. Someone aiding the creation of an infringing work can be acting tortuously in "contributory infringement" and hence linking to a website could, strictly, be tortuous. Hmm, I hadn't thought about it from that point of view, but I think you're right. It seems like clear evidence that your copyright law (I as…

>"I assume you're in the US"

UK actually at present - US has some things better than us and some worse. It all could do with a good going over though!

Post reply on HN