Re: [twitter-dev] Twitter Development platform - A Rant
Umashankar Das
The current debate on Nokia and it's viability has been a prime-example of that. It is well-known that Java became popular despite efforts by various forces to discourage it's widespread usage .
Now, Twitter and it's API groups claim that, they, are putting artificial limits of rates to ensure proper delivery for regular service. Someone has not studied the history of path-breaking products up there. If the method here is to discourage developers to invest in twitter to create applications, it is not the best idea. There is a lot of talk in market ccts that Twitter is looking at different ways to monetise it's huge userbase. Somehow, the recent actions leads us to believe , that, with that kind of focus you guys are losing track of what twitter can actually become.
Twitter is already a successful product. It was a great idea which has changed the way people communicate with each other. Today, revolutions happen in twitter. With all these restrictive views and thought processes, one believes that, perhaps, It does not think like a startup anymore. Guys, your role model should be facebook. Even today, they think like startups. It was facebook's open development platform which was the driver to more users in it's 2nd phase of expansion.
We in our company view twitter as much larger than micro-blogging product. We believe it can symbolise the next WWW. Our product lines are designed on that. We dont have whitelisting and we were not planning to apply for it. We were willing to work around the different restrictions and come up with innovative solutions. These restrictions have increased resource requirements for the product we're building, but out here, we're happy about it. They've got us people who have innovation which is essential to a successful startup.
Somehow, the recent statements make me believe that twitter is losing it here. I come from the domain of server technologies. I completely understand that QoS (Quality of service) is a big issue. But, in times of peak load, twitter still does go down. By putting artificial limits on querying twitter, you're not allowing creation of the next generation of products. All these limits indirectly makes one believe that twitter does not want more users. That seems very contradictory for a company whose valuation of 8-10 Billion USD is a function of it's userbase. If I were given the charge of solving this problem i would add new servers . Our server requirement[dedicated] costs us about 140 USD permonth/perserver. It has been said that you guys use Cassandra. Given Cassandra's performance replication efficencies are fairly decent. A fleet of 10 servers should be enough to handle things for you guys for the near future. That will be less than what you guys pay a developer.
I'm sure lots of users will be happy to pay for this too. Although, the real trick is to be free.
The future is about being open. Facebook was a means of bringing your social experience online. But, there was a concept of exclusivity to it. Twitter was diametrically opposite. It allowed people to talk to anybody. Politically speaking, it was a very democratic experience.
I would like to see twitter bigger than Google someday. Blocking development platforms is not the way to go.
Regards
Umashankar Das
The behavior on this group has changed significantly since Ryan
finally admitted that Whitelisting no longer exists. I've never seen
anyone discuss methods of getting around TOS before, well there was
Edward H., and we saw what happened to him. Now there are free flowing
discussions of MTurk and other tricks to go way beyond the rate
limits. I think this is great. Frankly, Twitter has done a good job of
offering free resources to devs, which I thank them for, but there was
way too much fear before. Now there are no extra benefits that can be
given and withdrawn "on a case by case basis." Boy do I hate that
phrase. Of course, they can ban people from this list, but maybe the
irony of Twitter blocking free speech on their own forum may restrain
that urge in the future.
Personally, I've treated Whitelisting like Social Security. It ain't
going to be there when I need it. That has turned out to be a winning
strategy. I don't really violate TOS, since I'm not as spammer, but I
have never tried building anything that would fail if Twitter didn't
give me Whitelisting "after it got into production", which BTW was the
most disrespectful thing I've seen from a platform vendor. Everyone
should assume that you need to use what is there by default, and
always be ready with a workaround if that gets taken away. My gut
tells me that things will get worse before they get better. Twitter HQ
will be under huge pressure to make money before the IPO, and we are
likely to get some of the cuts. The inevitable "they are parasites
leeching off of us" will surface. Anyone here old enough to remember
Ed Esber? But in the long run, I've never seen a global phenomenon
like Twitter, so I'm in it for the next 10 years at least. Then I can
retire.
Let's keep the discussion open guys. They've already taken away the
most important thing you wanted. Now we can build with our eyes open.
And don't be afraid to speak up. This is Twitter. Revolutions happen
here.
Adam Green
Twitter API Consultant and Trainer
http://140dev.com
@140dev
--
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: http://groups.google.com/group/twitter-development-talk
Andrew W. Donoho
On Feb 13, 2011, at 08:53 , Umashankar Das wrote:
Now, Twitter and it's API groups claim that, they, are putting artificial limits of rates to ensure proper delivery for regular service.
Mr. Das,
While you make many interesting points in your "rant", I think many of them are conjecture and opinion. As reasonable people can disagree about opinions, I've edited them out of my reply. I wish to focus on some unambiguous issues. We each have to make our business bets with respect to the Twitter platform. (I speak as the developer of ch@tter™, an iPad Twitter client. In many ways, Twitter destroyed my business opportunity when they purchased Tweetie and made it free. I mention this for context and not as a cause to rant at Twitter. I'm making plenty of money as a result of building ch@tter™. The iOS consulting business is very healthy.)
It is clear from this thread that many developers made, perhaps unwisely, product plans based on Twitter's continued support for white listing. In my case as a client developer, the increase of my API count from 150/hour to 350/hour due to moving to OAuth totally removed my need for white listing. If user streams was supported, I could easily live with 150/hour limit. If they would stand behind their user streams API, I would switch to it immediately. (Beta status is not, frankly, good enough. If they cannot make a commitment to their new API, why should I? By my count, user streams has been in beta for almost 6 months.)
Changing a platform's API is hard. Twitter is discovering this the hard way. Every developer has an investment they would like to preserve in the status quo. That said, Twitter's API evolution practices, presumably approved by their CTO, Mr. Sarver, are not, in my opinion, helping their partners grow with Twitter. That they are turning off white listing while not having yet made a production commitment to user streams, is a great example of an evolutionary stumble. That they haven't announced any other methods of enhancing Twitter's ability to scale while supporting functionality enabled by the large white lists is an oversight. The outrage expressed in this thread is good, unambiguous evidence of the stumble.
Another example is the closed roll-out of promoted tweets. I think every third party app developer would love to find a way to further monetize their Twitter application. Twitter did announce that they would find a way to allow their developer partners to participate with the promoted tweets program. That has not yet happened. Currently, as Twitter has made a floor price of $0.00 for iOS apps, I have to resort to Apple's iAds to capture revenue from my labors. I don't mind but it does cut my other market-making partner, Twitter, out of the revenue stream. As it reduces my revenue opportunities, I think this is sub-optimal. I win when my partner wins.
A third example is the annotation feature. I am sure all of us could find an excellent use for annotations. I have many ideas on how to use them. But I cannot.
Overall, everyone needs to remember that we are dealing with a company that publicly claims to not yet be trying to capture revenue from their platform. We are seeing from their experiments the collateral damage. Rolling with the punches is painful. That is the cost of trying to access the almost 200 million Twitter users.
____________________________________
Andrew W. Donoho
Donoho Design Group, L.L.C.
a...@DDG.com, +1 (512) 750-7596
Knowing is not enough; we must apply.
Willing is not enough; we must do.
-- Johann Wolfgang von Goethe
M. Edward (Ed) Borasky
<andrew...@gmail.com> wrote:
> It is clear from this thread that many developers made, perhaps
> unwisely, product plans based on Twitter's continued support for
> white
> listing. In my case as a client developer, the increase of my API
> count from 150/hour to 350/hour due to moving to OAuth totally
> removed
> my need for white listing. If user streams was supported, I could
> easily live with 150/hour limit. If they would stand behind their
> user
> streams API, I would switch to it immediately. (Beta status is not,
> frankly, good enough. If they cannot make a commitment to their new
> API, why should I? By my count, user streams has been in beta for
> almost 6 months.)
User Streams is in fact in production and has been for months. The only
restrictions on User Streams, other than what's documented in the
technical documentation, is that it is *only* for desktop *clients*, not
servers or mobile. I'm not sure where iPad fits in this spectrum, but
for sure an iPhone is mobile.
*Site* Streams is designed for servers and it is still in beta. Perhaps
you need to be pitching your idea to Twitter and adapting your service
to Site Streams if it's a server-backed app, which I'm guessing an
iPhone/iPad app would be.
> That said, Twitter's API evolution
> practices, presumably approved by their CTO, Mr. Sarver, are not, in
> my opinion, helping their partners grow with Twitter.
[snip]
> Another example is the closed roll-out of promoted tweets. I think
> every third party app developer would love to find a way to further
> monetize their Twitter application. Twitter did announce that they
> would find a way to allow their developer partners to participate
> with
> the promoted tweets program. That has not yet happened. Currently, as
> Twitter has made a floor price of $0.00 for iOS apps, I have to
> resort
> to Apple's iAds to capture revenue from my labors. I don't mind but
> it
> does cut my other market-making partner, Twitter, out of the revenue
> stream. As it reduces my revenue opportunities, I think this is
> sub-optimal. I win when my partner wins.
The key word in this rant is "partner". A *partner* is, IMHO, someone
who has a *formal* partnership arrangement. Sure, there's a certain
formality when you accept Twitter's TOS, but I think if you want to use
Site Streams or Promoted content, you should be negotiating as a
business with Twitter as a business. What's in it for Twitter?
Twitter has built a powerful brand. I was there in early 2007 when the
vast majority of pundits predicted that it would go nowhere - that it
was just a bunch of Ruby hackers with too much time on their hands, that
it would destroy flow, etc. It's now one of the top ten sites world wide
according to Alexa. If you want to be a "partner" with Twitter, *you*
are the one who needs to have something to offer *them* IMHO.
[snip]
> Overall, everyone needs to remember that we are dealing with a
> company that publicly claims to not yet be trying to capture revenue
> from their platform.
I seem to have missed that claim. As far as I know, they *are* trying
to capture revenue through a combination of Promoted Accounts, Tweets
and Trends with bundled analytics and data licensing.
> What do I want? I want a better developer experience. Both Apple and
> Microsoft show what a good experience can be. I want user streams, a
> promoted tweet API and annotations. I hope Twitter can deliver these
> technical features to enable new business opportunities for
> themselves
> and the Twitter app ecosystem. Myself included.
I think you have User Streams, though it may not be suitable for your
specific application. You may be eligible to get in the Site Streams
beta, although I'm guessing that was invite-only. You can always ask -
as a business negotiating a partnership with another business.
We'll have to wait and see about the Promoted products. Advertising
sales is a fiercely competitive business and it's not something I
personally want to deal with at the moment.
Annotations? That was definitely a case where Twitter's reach seems to
have exceeded its grasp. The story I've heard is that there are people
in Twitter hacking away on it but the priorities do get adjusted
according to the demands of the marketplace. If it could be a
breakthrough spam killer, I think they'd push it front and center in a
big hurry. ;-)
--
http://twitter.com/znmeb http://borasky-research.net
"A mathematician is a device for turning coffee into theorems." -- Paul
Erdős
Adam Green
big point. I think you are thinking too small when you say " If you
want to be a "partner" with Twitter, *you* are the one who needs to
to Twitter. 10,000 devs is a labor force. 100,000 devs is a market
that protects Twitter from *any* competitor, including Google. We are
all partners, because we all make money. You look old enough to
remember dBASE. That was a huge labor force that protected Ashton-Tate
for years when they had a product with technical limitations. Sound
familiar? Corporations and government agencies used dBASE not because
it was *best*, but because they could find many qualified developers.
Ashton-Tate started attacking their developers in 1988, when they were
one of the top 5 software companies. They were out of business 3 years
later.
If Twitter wants to be embedded into the infrastructure of
corporations around the world, they must have outside developers. If
they want it to be a cool toy for the Kardashians and Justin Beiber to
amuse their fans. They don't need us at all. It is their choice.
> --
> Twitter developer documentation and resources: http://dev.twitter.com/doc
> API updates via Twitter: http://twitter.com/twitterapi
> Issues/Enhancements Tracker:
> http://code.google.com/p/twitter-api/issues/list
> Change your membership to this group:
> http://groups.google.com/group/twitter-development-talk
>
--
Umashankar Das
Thank you for such a measured response. I'm not looking at the immediate action of stopping whitelisting by twitter. I must add, that, if a certain feature requires an application process, I had doubts in my ability to convince twitter, that, I really deserve to be in the list.
I believe I've also mentioned in my mail ,that , we, as a team/company, have already made arrangements which does not require us to use whitelisting. Perhaps, since, my initial mail was a rant, but, not a structured 'reply' it got lost in whatever I said.
What I say is essentially this. Twitter seems to be discouraging developers whose products have the potential who increasing their userbase. It almost as If I'm hearing that, 'WE DONT WANT ANYMORE USERS' . We're tired of managing the existing number.
I don't disagree that nobody should make plans on features which have no guarantees [whitelisting], that, they will continue in future. But, by putting such limits isn't Twitter running the risk of becoming just another product.
The creation of the facebook development platform was responsible in spawning a market leader in gaming field i.e ZYNGA.But, These limits by twitter dont allow someone to think about a new idea. He will stop thinking that ; what if I get traffic ? Anyhow, twitter will block me if I cross their rate limit. How can an innovator think in such an environment?
I'm not sure if you find this as conjecture. This , to me , is a logical line of reasoning.
Anyhow,
We ,as a team , have planned on working with a worst case scenario. twitter's search API is very important to us. We never planned for whitelisting, we dont need it. The only worry is if tomorrow, (hopefully), if we generate traffic from our product, will twitter just stop the 'SEARCH API' . We need to know that.
Clarity in this will go a long way.
I hope I haven't offended anybody here. These are views , since, we as a team , feel that we can create something path-breaking using twitter as our backend resource. We will only help twitter that way.
Thanks & Regards
Umashankar Das
Dewald Pretorius
as if you're not aware yet of the [mostly unofficially documented]
limits and constraints [mostly just alluded to by John Kalucki and
others through their exhortations on this list for people to rather
use the Streaming API] on using the Search API.
If your application is based on the assumption that you can throw an
unlimited [or even just a reasonably elevated] number of queries at
the Search API as and when your app needs scaling or hits a volume
spike, you will be wise to rethink your approach.
Andrew W. Donoho
On Feb 13, 2011, at 13:04 , M. Edward (Ed) Borasky wrote:
> On Sun, 13 Feb 2011 12:18:11 -0600, "Andrew W. Donoho" <andrew...@gmail.com> wrote:
>> It is clear from this thread that many developers made, perhaps
>> unwisely, product plans based on Twitter's continued support for white
>> listing. In my case as a client developer, the increase of my API
>> count from 150/hour to 350/hour due to moving to OAuth totally removed
>> my need for white listing. If user streams was supported, I could
>> easily live with 150/hour limit. If they would stand behind their user
>> streams API, I would switch to it immediately. (Beta status is not,
>> frankly, good enough. If they cannot make a commitment to their new
>> API, why should I? By my count, user streams has been in beta for
>> almost 6 months.)
>
> User Streams is in fact in production and has been for months. The only restrictions on User Streams, other than what's documented in the technical documentation, is that it is *only* for desktop *clients*, not servers or mobile. I'm not sure where iPad fits in this spectrum, but for sure an iPhone is mobile.
>
> *Site* Streams is designed for servers and it is still in beta. Perhaps you need to be pitching your idea to Twitter and adapting your service to Site Streams if it's a server-backed app, which I'm guessing an iPhone/iPad app would be.
Mr. Borasky,
Any API where the very first line in the schema section says it is "subject to change" is not, in fact, in production. Now it may be due to sloppy documentation but I doubt it. Twitter's documentation process has become much better.
IOW, user streams is in beta. If it is production worthy, then Twitter should commit to it. This API, until they version the route, should no longer be subject to change.
>> That said, Twitter's API evolution
>> practices, presumably approved by their CTO, Mr. Sarver, are not, in
>> my opinion, helping their partners grow with Twitter.
>
> [snip]
>
>> Another example is the closed roll-out of promoted tweets. I think
>> every third party app developer would love to find a way to further
>> monetize their Twitter application. Twitter did announce that they
>> would find a way to allow their developer partners to participate with
>> the promoted tweets program. That has not yet happened. Currently, as
>> Twitter has made a floor price of $0.00 for iOS apps, I have to resort
>> to Apple's iAds to capture revenue from my labors. I don't mind but it
>> does cut my other market-making partner, Twitter, out of the revenue
>> stream. As it reduces my revenue opportunities, I think this is
>> sub-optimal. I win when my partner wins.
>
> The key word in this rant is "partner".
Let me be very clear. The term "rant" was applied by the opening poster.
I am pointing out what I think are concrete issues which we all, as business partners with Twitter should consider. That is not a rant.
> A *partner* is, IMHO, someone who has a *formal* partnership arrangement. Sure, there's a certain formality when you accept Twitter's TOS, but I think if you want to use Site Streams or Promoted content, you should be negotiating as a business with Twitter as a business. What's in it for Twitter?
The ToS is a contract. I've also based my business on Twitter's APIs. I view the APIs and other aspects of the relationship I have with Twitter to effectively be a partnership. Twitter, as the senior partner in this relationship, barely knows I exist.
As to what's in it for Twitter? They have support a public API because it makes their product better. And they do want people like me making their and my product better.
> Twitter has built a powerful brand. I was there in early 2007 when the vast majority of pundits predicted that it would go nowhere - that it was just a bunch of Ruby hackers with too much time on their hands, that it would destroy flow, etc. It's now one of the top ten sites world wide according to Alexa. If you want to be a "partner" with Twitter, *you* are the one who needs to have something to offer *them* IMHO.
And, I believe I do. I suspect that you believe you offer them value and vice versa or you would not waste your time on this list. This email and those of other developers do improve their product. Every bug discovered helps their product.
Are we suckers? Perhaps. Overall, I have profited from this arrangement. I wish Twitter well and I hope they improve their relationship with the developer community.
>
> [snip]
>
>> Overall, everyone needs to remember that we are dealing with a
>> company that publicly claims to not yet be trying to capture revenue
>> from their platform.
>
> I seem to have missed that claim. As far as I know, they *are* trying to capture revenue through a combination of Promoted Accounts, Tweets and Trends with bundled analytics and data licensing.
"Stone, reacting to a question from TechEye, said that there was still so much the company wanted to do, including proving that it had a viable financial and sales profile."
OK, so I misremembered the quote. What is clear is that they haven't even proven to themselves that they have a business.
>> What do I want? I want a better developer experience. Both Apple and
>> Microsoft show what a good experience can be. I want user streams, a
>> promoted tweet API and annotations. I hope Twitter can deliver these
>> technical features to enable new business opportunities for themselves
>> and the Twitter app ecosystem. Myself included.
>
> I think you have User Streams, though it may not be suitable for your specific application. You may be eligible to get in the Site Streams beta, although I'm guessing that was invite-only. You can always ask - as a business negotiating a partnership with another business.
As I say above, user streams is not in production. Yes, they just have to remove a single sentence and declare it done but they haven't done that. Mature engineering organizations do promote items formally to production. As my app does not need user streams, I can wait until they are ready to commit to it. It is a modest extension of my code. As white listing was one of those "nether" commitments, I am not surprised that it has been withdrawn. Site streams may be what all of the white list dependent folks need. But, Twitter canceled white listing without making site streams widely available. That is an example of an API evolutionary stumble.
> We'll have to wait and see about the Promoted products. Advertising sales is a fiercely competitive business and it's not something I personally want to deal with at the moment.
The point of Twitter encouraging me to show promoted tweets, as they said was their goal, is to enhance each or our revenue streams. They do the sales and I do the display. Seems pretty easy to me.
> Annotations? That was definitely a case where Twitter's reach seems to have exceeded its grasp. The story I've heard is that there are people in Twitter hacking away on it but the priorities do get adjusted according to the demands of the marketplace. If it could be a breakthrough spam killer, I think they'd push it front and center in a big hurry. ;-)
A mature engineering organization backs away from things they aren't going to do. Fortunately, my business does not depend upon annotations. I could do interesting things with annotations.
Let me be very clear. Twitter is a great resource. Could they be a great company? Perhaps. But to be a great company means you have great partners. If you look at Apple, Microsoft, IBM and Oracle, great companies all, they have significant partner ecosystems. Those partners make those companies great -- and vice versa.
Anon,
Andrew
____________________________________
Andrew W. Donoho
Donoho Design Group, L.L.C.
a...@DDG.com, +1 (512) 750-7596
"We did not come to fear the future.
We came here to shape it."
-- President Barack Obama, Sept. 2009
M. Edward (Ed) Borasky
wrote:
> Edward, I'm going to jump in on the partner issue, since that is my
> big point. I think you are thinking too small when you say " If you
> want to be a "partner" with Twitter, *you* are the one who needs to
> have something to offer *them* IMHO." One dev is very small compared
> to Twitter. 10,000 devs is a labor force. 100,000 devs is a market
> that protects Twitter from *any* competitor, including Google. We are
> all partners, because we all make money. You look old enough to
> remember dBASE. That was a huge labor force that protected
> Ashton-Tate
> for years when they had a product with technical limitations. Sound
> familiar? Corporations and government agencies used dBASE not because
> it was *best*, but because they could find many qualified developers.
> Ashton-Tate started attacking their developers in 1988, when they
> were
> one of the top 5 software companies. They were out of business 3
> years
> later.
>
> If Twitter wants to be embedded into the infrastructure of
> corporations around the world, they must have outside developers. If
> they want it to be a cool toy for the Kardashians and Justin Beiber
> to
> amuse their fans. They don't need us at all. It is their choice.
Well, I'm old enough but I was doing something radically different from
Ashton-Tate at the time. This whole thread is starting to sound eerily
similar to last year, when Fred Wilson made the infamous "filling holes"
blog post, followed by Twitter buying Tweetie, followed by Chirp. I'd be
surprised if the *Twitter* ecosystem could support 10,000 independent
developers - they'd self-organize into businesses with some sort of
power law size distribution, where the largest such business is Twitter
itself.
I don't know that Twitter "wants to be embedded into the infrastructure
of corporations." It seems to me that Twitter is unique and not at all
suited to intra-enterprise communications. Besides, there are dozens of
enterprise software platforms that can do everything Twitter can do
except talk to the hundreds of millions of Twitter users in real-time.
;-)
Maybe I am thinking too small, but then again, people aren't coming to
*me* with problems big enough to require whitelisting, or for that
matter Cassandra, or MapReduce, or sending thousands of DMs a day. Even
if they did, there's no way I could compete with Twitter. I really
should save this for my blog - it's been a while since I wrote a post
about Twitter, and that's what my search analytics tell me people read
there. ;-)
Adam Green
> Ashton-Tate at the time. This whole thread is starting to sound eerily
> similar to last year, when Fred Wilson made the infamous "filling holes"
> blog post, followed by Twitter buying Tweetie, followed by Chirp. I'd be
> surprised if the *Twitter* ecosystem could support 10,000 independent
> developers - they'd self-organize into businesses with some sort of power
> law size distribution, where the largest such business is Twitter itself.
>
> I don't know that Twitter "wants to be embedded into the infrastructure of
> corporations." It seems to me that Twitter is unique and not at all suited
> to intra-enterprise communications. Besides, there are dozens of enterprise
> software platforms that can do everything Twitter can do except talk to the
> hundreds of millions of Twitter users in real-time. ;-)
>
> Maybe I am thinking too small, but then again, people aren't coming to *me*
> with problems big enough to require whitelisting, or for that matter
> Cassandra, or MapReduce, or sending thousands of DMs a day. Even if they
> did, there's no way I could compete with Twitter. I really should save this
> for my blog - it's been a while since I wrote a post about Twitter, and
> that's what my search analytics tell me people read there. ;-)
>
developer. It could mean someone who builds a web or mobile app and
tries to monetize it. That would be limited. I think it also means all
the consultants and in-house programmers who integrate Twitter into
existing websites and businesses. As I started responding CNN ran a
big button on the screen telling people to try their Twitter
integration on their website. I think that was built by a developer,
not Twitter HQ. Multiply that by every TV show, radio program,
newspaper, magazine, movie, real estate office, hospital, retailer,
you get the point. There are way more than 10,000 programmers who work
on websites and mobile apps around the world. They are all possible
Twitter developers, among other tasks they did. Too big an idea?
Maybe, but with the right assistance from Twitter, there would be
enough developers that when a competitor comes along Twitter would
have a base that would make it hard to switch. That is what we offer
them.
I have an idea. Why doesn't Twitter hire a developer relations person?
Not a support person. Matt and Taylor do a good job of technical
support. I appreciate what they do. I mean someone who could run a
developer program. I haven't seen someone like that yet. Could some of
the $200 million pay that salary?
I look forward to your blog post on this, Edward.
M. Edward (Ed) Borasky
wrote:
> Good points. I think the basic confusion is the definition of
> developer. It could mean someone who builds a web or mobile app and
> tries to monetize it. That would be limited. I think it also means
> all
> the consultants and in-house programmers who integrate Twitter into
> existing websites and businesses. As I started responding CNN ran a
> big button on the screen telling people to try their Twitter
> integration on their website. I think that was built by a developer,
> not Twitter HQ.
My impression is that this is exactly the sort of thing @anywhere was
designed to do - make it possible for a CNN or even the Original Coffee
Brake to incorporate Twitter into their web site with a budget of, say,
8 hours of HTML editing time. ;-) I haven't kept up with how well
@anywhere is fulfilling that promise, though. I ran it for a long time
on my blog but shut it down because the trips to Twitter's servers were
slowing down page loads. I should probably revisit that now that I'm
starting to get traffic again.
> Multiply that by every TV show, radio program,
> newspaper, magazine, movie, real estate office, hospital, retailer,
> you get the point. There are way more than 10,000 programmers who
> work
> on websites and mobile apps around the world. They are all possible
> Twitter developers, among other tasks they did.
I don't know about the rest of the world, but here in PDX, the skills
that are in huge demand are HTML5/CSS3/JavaScript and user interface
design. We've got a small collection of people who "do stuff with
Twitter", but you don't see help wanted ads for Twitter API coding -
that's something people do in their "spare time". Most folks use Twitter
the old-fashioned way - from the web app or from a client - or license a
monitoring platform that talks to Twitter and Facebook.
> Too big an idea?
> Maybe, but with the right assistance from Twitter, there would be
> enough developers that when a competitor comes along Twitter would
> have a base that would make it hard to switch. That is what we offer
> them.
I think it would be harder for a competitor to get Twitter's millions
of active subscribers than to get thousands of developers. ;-) I was
just looking at the Alexa statistics - Twitter is in 9th place
world-wide now.
Who's ahead of us? Twitter is just below the huge Chinese site
Baidu.com. Neither Facebook nor Twitter is active in China, although I
have seen accounts claiming to be from Guangzhou. Next up the ladder is
Wikipedia. In short, it's been a long climb since March of 2006 to get
there, and there's a lot of power above Twitter -
Google/Youtube/Blogger, Yahoo, Microsoft, Facebook, Wikipedia and
Baidu.com. Twitter is, as they say, running with the big dogs. And we're
ahead of Aol. ;-)
> I have an idea. Why doesn't Twitter hire a developer relations
> person?
> Not a support person. Matt and Taylor do a good job of technical
> support. I appreciate what they do. I mean someone who could run a
> developer program. I haven't seen someone like that yet. Could some
> of
> the $200 million pay that salary?
Maybe again I'm thinking small, but I have yet to run up against
anything that Twitter did that seriously impacted me. Twitter's not like
Microsoft, Android or Apple where you need a huge standardized SDK /
MSDN-like library. The one thing I'd want as an independent developer
would be some kind of keyword tools along the lines of what Google
provides for webmasters. I can easily determine what people tweet about
but I *can't* determine what they search for. Oh, yeah - a Streaming
endpoint that delivers the overall tweets per minute every minute, so I
can draw pretty graphs in real time like Carolyn Penner did on the
Twitter blog.
http://blog.twitter.com/2011/02/superbowl.html