I'm always a little miffed when I see natural language being the input, because now in order to use the tech I need to also play the "guess the verb" games.
This is especially bad in Wolfram Alpha.... I can clearly see that all parts of a query are supported but I have to get the NLP part to parse my sentences right. Though there is Mathematica I guess....
Not that I dislike all natural language stuff (far from it) but I'd love to have access to an AST builder too
I generally agree, but there are some cases where I've found natural language actually is the right interface. And complex interfaces with lots of touch points but a simple underlying principle seems to be that sweet spot.
For example, I exclusively use Siri to set reminders for times and places, because to do so using a typical touch interface is much more painful.
Whether or not this CAP idea will meet the bar is an open question for me.
I find it telling that your comparison is to touch interfaces. The (unintentional?) inference being that only touch interfaces are painful enough that current generation NLP can compete with them.
I use keyboard and mouse for setting reminders. The app I really miss now I don't have a smartphone is google maps.
If you don't have a smartphone it will be difficult for you to do this, but try creating a time- or location-based reminder using a keyboard interface. And I'm generously assuming that you're going to run this test at a computer, and not directly after bumping into someone in the hallway who reminds you about something you need to do.
This is a bit off-topic from the CAP thing, where I think it's reasonable assume that you are at a computer.
I'd like a language with "natural language" as a first class citizen of a language so instead of trying to name methods, you break down a description of code into paragraphs in english, then you could have a parser that attaches lambdas to those paragraphs.
you would also need the editor to know to display a separate implementation file for each paragraph.
I get where it comes form, and it makes sense, but it's long and will inevitably be shorterned to IFTTT which is just awkward.
An acronym/initialism is better if it can be pronounced like a word. CAP is one of them, and thus I feel it's better than IFTTT even though IFTTT's full name makes more sense and is 'good'.
In terms of searchability, "IFTTT" and "Zapier" beat "CAP" by a mile because the latter is a common word. So I can search for recipes with "IFTTT" or "Zapier" but in this case I have to type the full expansion "conditional action programmer."
Whether or not a name like IFTTT is awkward, and therefore bad in some sense, is subjective. Personally I happen to like the awkwardness.
> Westm. Papers 1 Nov. 107 Who has not suffered when he has played correctly second hand..from his partner assuming that there has been an ask for trumps?
> 2014 D. Chapman Ascendant lxxiii. 301 A plane? Full of people?.. That is an enormous ask.
When used liked that it's synonymous with request. If heard it often used as "that's a big ask". Some example usages[1] since you seem a bit skeptical.
I'm referring to Microsoft's moves in the past year to release their compiler and other parts of the .NET toolchain (and Xamarin) as MIT-licensed open-source projects (moves the company wasn't forced to make). Even though the language itself has been an open standard for some time, the C# ecosystem was very different a few years ago than it is in 2016.
Microsoft even made the C# language specification ECMA and ISO standards. The ECMA process started shortly after the first commercial release. I'm pretty sure if they had just talked to the Microsoft guys, they would have realized everything would be fine.
Not really relevant to this particular conversation. C# was commercially released in July 2000. The ECMA standardization committee was founded in September of the same year. Writing began in January 2003, and the standard officially adopted in June 2006. This is all described in the standard:
I didn't say C# was released after this time, I said Microsoft is a different beast now. They've made a significant swing towards open source and embracing the OSS community. Back in 2005 they still very much thought they ruled the world and could get away with anything.
Your opinion on 2005 Microsoft is great and all, but I still don't see its application to this exact conversation. From 2006 and on, anyone could implement a C# compiler and runtime from the ECMA standard with absolutely no barriers from Microsoft. This was the result of a process that started in 2000, a few months after the official release of the language.
I think it is relevant. The Microsoft of that time (and it is debatable how much that changed) is not one company on which you'd rely on for a cooperation in such a project. There is no telling which tricks MS might have used to make money out of it – the same way Oracle's API copyright bullshit was a trick to make money. Don't forget that Microsoft uses patents to extort a share of many android phones sold, see http://uk.businessinsider.com/microsoft-android-patent-licen.... I thought that had stopped by now, but that article is from april…
So does Google, so does Apple and so does Samsung. If you don't protect your rights they will be taken away. The mobile market is in a state of mutually assured destruction. The fault lies with our broken patent system and all the players in that market: not Microsoft alone.
This is correct. The Microsoft apps on android are phenomenal. I think it speaks volumes about the android toolchain and development scope when a direct competitor can make better apps than for their own platform.
I was going to comment on the fact that Microsoft has released a lot of their OWN tooling for Android. But I wanted to issue a quick correction to your comment: Android development hasn't been Eclipse based in years. Google went over towards encouraging IntelliJ, and Android Studio is based on it.
The very next sentence is "We need your help to continue to improve natural language understanding." Maybe a joke? Or maybe they wrote this on a cell phone?
I find it very amusing when people name other brands/apps etc on their homepage especially those where you don't have any business connection.
Case in point here is naming Pocket App on the homepage prominently. It gives on impression that Pocket is some big successful app but in reality has merely 22M users (active users must be lot lower).
I hope I am right to assume that MS has nothing to do with Pocket app.
Because of the post the other day about Bongard problems [1] I can't help but wonder if it is possible to use that author's algorithms to automatically detect the kinds of conditional patterns that this software wants to automate.
Anyone know what the "HTTP" integration is? If I could get this service to make an (authenticated?) web request I could make it hit a service on a box at home to turn some other stuff on.
Ability to turn my Xbox One on from a programmable button would be great, considering the Xbox Live team took Wake On Lan and turned it into something requiring a Microsoft-signed cert...
When I sign in with my outlook.com email address it asks for permission for the "Luis-Actions" app to read my email, among other permissions. It's probably Microsoft, but I wish they would have made the signup experience look a little more trustworthy.
Marketing is even worse at naming things! You don't see programmers having 2 weeks of meetings to come up with a terrible, terrible name and still feeling good about themselves afterwards.
This is especially bad in Wolfram Alpha.... I can clearly see that all parts of a query are supported but I have to get the NLP part to parse my sentences right. Though there is Mathematica I guess....
Not that I dislike all natural language stuff (far from it) but I'd love to have access to an AST builder too