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

What Happens When A Twitter Client Hits The Token Limit

marco.org

Post reply on HN
Print view View original

51–60 of 188 posts

Re: What Happens When A Twitter Client Hits The Token Limit

#51

Post by macspoofing »

hnriot wrote: post #7

*The effective rule, therefore, is even simpler: “Don’t build anything for Twitter.” Exactly, that's precisely the message they wanted you to have. What's wrong with using the twitter.com on Windows8, do we really need a special client just for Windows 8? This is exactly what the web is supposed to do. I don't get anyone is surprised, it's Twitter's ecosystem and if you're duplicating their functionality then it's pe…

>What's wrong with using the twitter.com on Windows8

I dunno. Ask the 100,000+ people that wanted to use the client.

Re: What Happens When A Twitter Client Hits The Token Limit

#52

Post by xentronium »

tiles wrote: post #42
uptown wrote: post #38

Aside from pissing off Twitter, and possibly getting a cease and desist, what's preventing developers from building a translator that sits between Twitter's web layer and their native application client? Couldn't something be developed that loads Twitter into a hidden webview that's locally scraped for the purpose of re-display however the developer pleases on their client? This implementation wouldn't require the to…

Or automating the creation of a unique app token for each user: https://dev.twitter.com/apps What TOS does that violate? What if the user manually creates a token and inputs it? Could Twitter detect this even happening?

I'd say that sudden explosion in one user application is very suspicious. This is easily countered by premoderation of applications, for example.

Re: What Happens When A Twitter Client Hits The Token Limit

#53

Post by wccrawford »

artursapek wrote: post #35
quotemstr wrote: post #21

Earlier quoted context omitted.

I'm sure Twitter would block bots doing unuserlike things. But isn't the point of a Twitter client to do userlike things?

I suppose, and I suppose if the client used a user's IP they couldn't block them the same way. But wouldn't a client that scrapes the full HTML to get the data you want be horribly inefficient, and hard to maintain? API's are all the rage because they give you just what you want. Think about bandwidth.

Having written scrapers in the past, I can tell you that changes are usually pretty easy to deal with. And if you write your scraper nicely, many small changes won't even matter.

Since the client runs on the user's machine, performance isn't really an issue, either. (Trying to do it for all users on a server is a different story.)

APIs are indeed "all the rage" because they're easier and don't change as much. But when given no other option... You do what you have to.

Re: What Happens When A Twitter Client Hits The Token Limit

#54

Post by wccrawford »

ghshephard wrote: post #22
hnriot wrote: post #9

Earlier quoted context omitted.

Unless you work for Twitter, I'd say you're right, and rightly so. Don't depend on someone else's API, how many times must history repeat itself before this message sinks in.

That's an unreasonable position to take. Almost all third-party software products ever created rely on some other third-party APIs, and to suggest "Don't depend on someone else's API" as a strategy would be akin to saying "Don't write software for a living." The challenge is to determine, in advance, if the platform you are writing for will maintain some reasonable level of access to the API, both now, and in the fut…

Perhaps "don't depend on someone else's API unless you're paying for it" is a better attitude. Depending on someone else to keep giving you something for free is problematic at best.

Re: What Happens When A Twitter Client Hits The Token Limit

#55

Post by spoonmonkeyuk »

quotemstr wrote: post #23
eli wrote: post #19

Earlier quoted context omitted.

To what end? You're going to base a business on scraping twitter.com? That sounds like a much worse idea than trying to work within the API.

> You're going to base a business on scraping twitter.com? A business? Probably too high risk. A nice FOSS client, though? Why not?

Why not? Because of liability to law suits for breach of the terms of service or copyright violation. Third parties have been sued for scraping in the past.

Re: What Happens When A Twitter Client Hits The Token Limit

#56

Post by dannyr »

quotemstr wrote: post #8

You know, back before APIs were all the rage, people wrote clients for web services by scraping. Twitter really wouldn't be able to do anything about a Twitter client that pretended to be IE9.

This would only work for the public timeline. Most people would want to view their own timeline and that would be hard to since you'll have to create a browser instance for every user you have.

Re: What Happens When A Twitter Client Hits The Token Limit

#57

Post by bobzimuta »

barrkel wrote: post #20
hnriot wrote: post #9

Earlier quoted context omitted.

Unless you work for Twitter, I'd say you're right, and rightly so. Don't depend on someone else's API, how many times must history repeat itself before this message sinks in.

That's a ridiculous statement. Unless you are as vertically integrated as Apple, you can't avoid dependencies on someone else's APIs.

What's ridiculous is assuming you can make money off of a free data source without any inherent long term risk.

Re: What Happens When A Twitter Client Hits The Token Limit

#58

Post by Skroob »

wmf wrote: post #44
Skroob wrote: post #37

Earlier quoted context omitted.

Because their customers are advertisers, and their users are people who follow Ashton Kutcher and Justin Bieber. It's not the same service as it was when it started, when the API was important to them.

If Twitter charged to send tweets (like Facebook is now doing) then they wouldn't need to crack down on third-party clients.

But who wants to pay to tweet? With the business and culture they've built, they don't have a way to go back. They have to stay free for everyone, they have to show ads to make money, and they have to kill 3rd party apps to make sure the ads get seen their way.

If you want Twitter-that-was, ADN is your best bet at this point.

Re: What Happens When A Twitter Client Hits The Token Limit

#59

Post by TheCapn »

This is the inherent problem with building your product with dependencies on other products. You are tied to their system in such a way that your existence relies on the faith that they keep doing what they do and do not change to a system that blocks you or inhibits your functions.

I see the need/desire for hackers to make things useful and unique in a way that they envision things but you will never hear me apologize for my remarks on this subject. If you are building your business or product depending on someone else's and are not a form of contractual partner you can be kicked in the ass down the line and there's no accountability to you they owe.

Re: What Happens When A Twitter Client Hits The Token Limit

#60

Post by egypturnash »

hnriot wrote: post #7

*The effective rule, therefore, is even simpler: “Don’t build anything for Twitter.” Exactly, that's precisely the message they wanted you to have. What's wrong with using the twitter.com on Windows8, do we really need a special client just for Windows 8? This is exactly what the web is supposed to do. I don't get anyone is surprised, it's Twitter's ecosystem and if you're duplicating their functionality then it's pe…

There are things clients do much better than the website. Tweetbot, for instance, makes having multiple accounts a LOT easier than logging out and back in. Plus stuff like OS notifications, general OS integration, filtering ("oh god I'm sick of hearing about #obama and #romney, I already voted")… Clients definitely make Twitter more worthwhile for a user.

Sadly what makes Twitter more valuable for a user does not necessarily make the user more valuable for Twitter.

Post reply on HN