New Yahoo?

I suppose I should have an opinion on Yahoo bringing in Marissa Mayer to rescue it from (popular and profitable) non-entity.

Frankly I have no real insight. But here's the golden rule. You can't be a great technology company and a great media / content company at the same time. (Unless you are Nintendo, and then you're really just a great media company that understands technology well enough not to let it get the upper hand.)

Yahoo has, for as long as it's worth anyone remembering, been a media company. In fact it's been an OK media company, with lots of traffic and a respectable profit. It's been run as a media company, by media people. And its only real problem is that it's been judged and found wanting against the tech. giants of our age.


So, it's never cared about its geeks. It dabbles with technology in only a lacklustre way. And it's squandered some really great tech. acquisitions.

But suddenly, it wants to be a tech. company after all???


How is that going to work? Can Mayer do a 180 degree turn? Re-inspire the geeks who already work there? Bring in more talent? What will happen to all the media people? How will Mayer manage to keep them happy if she explicitly tries to turn Yahoo into a tech. culture? Or, will she fire them (as Marc Andreesen advises)? Done badly, this could be a disaster, with Yahoo ripping itself to pieces as it tries to slam into reverse.

Alternatively, if Mayer is really clever, she may figure out how to keep running Yahoo as a media company while doing something exciting with it. Like Nintendo, Yahoo *could* become a media company that understands technology well enough to use it properly.

Here are a few random ideas :

1) Yahoo is nowhere in mobile. It's never going to own a platform, a crucial service or make extraordinary hardware. Give it up. Yahoo mobile apps are simply a necessary cost of keeping its media channels in play.

BUT Yahoo, as a brand, *does* have a chance with TV. None of the tech. giants have successfully colonised TV yet. It's still an open field. Yahoo's brand, content, sales-force and in-house media know-how give it a shot at the smart TV market. It's not clear what plans Yahoo have for smart TV. Do they have an OS for it? Well, if not, here's a thought : partner with Ubuntu who do have a slick and robust operating system and interface for TV but are unlikely to make much headway with it by themselves.  A combination of Ubuntu's technological maturity with Yahoo's brand and audience (including its sports channel, partnerships with gaming companies etc.) could make up a compelling package for third-party TV manufacturers.

Seriously. No-one in the world really wants a "Google TV" or a "Microsoft TV". These brands have no connection with what anyone thinks of when they think of TV. The Yahoo brand has a great advantage here because people do recognise it as a *content* brand. A news, sport, entertainment portal. Which is what smart TV is aspiring to become.

2)  AOL made a play to be a great web media company in its acquisitions of Weblogs Inc., TechCrunch and The Huffington Post. Much has gone wrong, and some of the biggest and smartest personalities have left. Did that signal the idea was fundamentally flawed or just that the chemistry didn't work out that time? Might it be worth, for example, Yahoo buying Mahalo to bring Jason Calacanis into the fold?

3) More than that, online education is an area which is continuing to grow. Yahoo Answers is a massively popular site. But compared to Stack Overflow or Quora is looking decidedly old and unloved. What could Answers become if paired with Mahalo and put under Calacanis's influence and ambition?

(Aside : I would have suggested Yahoo go for Instructables too, but I just discovered that they were bought by Autodesk : a move I consider very sensible. That leaves another cute Squid-Labs offshoot : HowToons that could work in a Yahoo context.)

4) Perhaps what Yahoo! needs is a big statement. Something cool. Hardware, because hardware is what makes big statements. Apple had the iPhone. Microsoft had Kinect. Google is going with the glasses. Yahoo have the money to buy themselves into this game. So what about Parrot? Drones are coming, and Parrot have the leading consumer drone platform. AR Drones would get Yahoo into robotics. Into augmented reality. Would give it a noteworthy gaming / entertainment platform. Would inspire geek lust. Etc. And Parrot + a successful push into TV would open up an interesting home automation front.

5) Many people have commented on the sad decline of Flickr under Yahoo. Not sure there's much more to say, except, I'm very surprised that Flickr is not better represented on the Yahoo home-page. The Flickr blog has some wonderful pictures. I'm not sure why Yahoo don't promote Flickr and Flickr photographers on its front page.

6) OK. Showing my prejudice here. The only Yahoo service I actually use is Pipes. It's good, but again, its potential feels underdeveloped.

So let's be bold for a second. Pipes is the great, philosophical, something-or-other of our age. The shift we think of from web 1.0 to web 2.0. The move to social. To apps. The coming of Netocracy. These are all, fundamentally, about the shift from managing stocks to managing flows. Most people are now drowning in the deluge that floods through Twitter, G+ and the Facebook Wall. Even Email has never really been conquered. We are *desperate* for tools to manage our flows. And Pipes, one of Yahoo's nerdiest, most out-of-place, quirks of a service is potentially the solution. If it can, paradoxically, be both professionalised and made usable by the wider base of Yahoo members.

How could Pipes be professionalised?

a) The UI is cute. But programmers are never going to work in that kind of environment. Pipes should offer a plain text alternative. A simple, "little language" to define and edit piping networks. Have an editor in the browser. Allow the scripts to be accessed via Git (and so shared on GitHub). Provide syntax colourers for popular offline editors like Eclipse and Emacs. Make sure programmers can think of Pipes as quick to write and convenient to work with.

b) Pipes should be fast, and a cloud service. By which I mean, it should be possible to host a Piping network on a fast server and pay for it. Like Amazon AWS and Google App. Engine.

c) Pipes should talk to Twitter, Facebook, Google+, Status.net, Yammer, RSSCloud. Maybe even AMQP.


d) Most important though, Pipes should be integrated with Yahoo mail. You should be able to use it, inside Yahoo mail to filter your incoming messages, send auto-replies, create alerts and summaries etc.


e) You should be able to embed a Pipes network within an external site. In fact, it should be possible for a third-party to build a product such as RSSGraffiti on top of Pipes. And build it in an afternoon.


At the same time, how can Pipes be made more accessible?

a) The most popular rival to Pipes is probably http://ifttt.com/wtf . If I was running the Yahoo Pipes strategy I'd look seriously at whether we could buy it or hire its people. At the very least we'd be pitching Pipes as a rival to it and offering similar functionality on the same engine.

b) Remember the push into education I suggested? Pipes could gain a more child-friendly UI putting it in competition with MIT Scratch and Google Blockly. Promoted in a "learn to program" course, via Yahoo's own kids / study zone it could gain some traction in schools.


Does the data-flow model of Pipes lend itself to more general programming tasks? Sure if you look at the use of Max/MSP and PD for music or VVVV for graphics. And, I'd argue that, building piping networks is likely to be an increasingly important metaphor for programming in the worlds of the cloud and device swarm.


Imagine, for a second, Yahoo teaching kids to create their own VVVV-style eye-candy on large-screen Yahoo branded TVs through a Pipes-like UI as taught by a series of lessons on the study zone. Imagine those skills then being useful for mashing up websites, business process modelling, etc.

c) An iPad app. Multi-touch is a great interface to compose piping networks.

7) Here is an outstanding question I'm trying to make up my mind about. Does Yahoo need its own browser? A long time ago I thought it absurd that Microsoft wasted its money and energy on a browser, when the future belonged to web-standards. More recently we've seen how important it is for would-be giants to have their own browsers. They give you the chance to push web-standards the way you'd like them to go. Or at least vote for the tentative standards you'd really like to see. Of course, a Yahoo browser would be most likely built on WebKit, just as Apple's and Google's are. But it might still be worth having. 


Update : Bill Seitz comments : http://webseitz.fluxent.com/wiki/z2012-07-16-MayerNewYahooCeo

Overview of mashup tools

Paul Graham thinks Microsoft is Dead. Dave Winer disagrees. Sort of.

The caveats don't matter. Graham's greater point is compelling. Microsoft don't really seem to be making it as an Internet Playa. They've neither come up with any really compelling or exciting software on the web-as-a-platform in the last seven years nor have they bought any of the interesting new web 2.0 companies. In fact, the last interesting thing MS did was buy Hotmail which was about 10 years ago.

10 years??? !!!!

...

!!!!!!

In the meantime, it more or less looks like they've bet everything on Vista and lost, big time. The desktop operating system is a commodity. There's no (and will be no) interesting software that really needs Vista. Web served applications can run as easily on Unix. Office and Photoshop will run as easily on Macintosh. And browser-based software will run anywhere.

Microsoft may still have platforms that matter - though it's not easy to imagine where : ASP.NET essential to web-based apps? XBox beating Wii or PS3? Ray Ozzie, their great hope, seems to be missing in action, last seen a year ago speculating about his clipboard while Yahoo has Pipes up and running.

Graham's suggestions for restoring Microsoft's relevance aren't nearly as interesting as mine.

Now, of course, the funny thing is in my day job I use nothing but Microsoft software. I basically live in Windows and (and this may reflect my new status as a project managery kind of guy) Excel.

In fact, this is something I'm trying to think about more. Excel is a truly great piece of software; it's Microsoft's masterpiece. Word is an OK Word Processor. Access is A.N.Other database. Powerpoint is ... well frankly let's not go there.

But Excel is wonderful. It's the universal, "Swiss Army" desktop solution with dozens of little functional "blades". Want a "to-do" list? Excel. Want a status report? Excel? Want to do some calculations? Excel. Want to do some basic string processing? I write VBA macros for the same kinds of simple data crunching that I'd use Perl for in Unix. Want to make a couple of graphs and charts? Excel. Want to mock up some forms? Want to make tables of data and sort and filter them? You guessed it ...

And not only does Excel does all this, it makes it all pretty intuitive. Have a look at how they do Pivot Tables for an example of something pretty slick.

No-one else is even close. Not Google's online spreadsheet. Not Open Office's attempts at catching up. Not WikiCalc. Microsoft's advantage with Excel is undisputed. It's all theirs to throw away.

And what sucks most about Excel? The fact that people are always mailing spreadsheets around to each other and they have trouble keeping a single, up-to-date copy between them. What they need is Excel socialized. And where's socialized Excel? Caught up in turf-wars and lost behind a bunch of vague, confusing products like "SharePoint" and technologies like Excel Services.

Now, if I ran Microsoft, and I was worried about Microsoft being dead, I'd be making the most I could of Excel : pumping money and smart people and advertising into it, setting up skunk-works, hiring clever explainers to get simple messages out, as loudly and clearly as possible.

In particular I'd have :

• Excel Studio : a complete development environment for people to build new applications on top of the Excel engine or to compile spreadsheet-based prototypes into other pieces of software.


• A Social Excel : the Excel client would allow many people to work on a shared spreadsheet either via a central web-server, LAN server, or simply sync. multiple users together over P2P (imagine something like a Skype call working on one spreadsheet.)


• Excel Live : a free, central web-based server to set up groups sharing the same spreadsheet with (obviously) Wiki-like (WikiCalc-like) hyper-linking between spreadsheets


• Excel Express : a completely free-as-in-beer cut-down version of Excel that anyone could download and use to work on a shared spreadsheet. I'd want Excel Express to be as easily available and viral as Skype or Pando.

99% of the world's "semi-structured" data is not in Microformats but in tables in spreadsheets. And, Microsoft pretty much own that. But there's a huge demand (and opportunity) to put it all on the internet. Like I say, this is Microsoft's platform to lose.

Hmmm. Social Networking news ...

Cisco buying Tribe's tech. People are confused.

Meanwhile Ning seems to have decided to reinvent itself as ... erm ... Tribe : a generic build-your-own-social-network site.

Me, I'm confused too. And more than that, I'm disappointed by the Ning change. Taking a quick look at the site, it seems to have utterly disowned the previous "social development environment" incarnation of Ning, handing over that mantle to Yahoo Pipes.

OK, I half understand : maybe the business model wasn't working. Not everyone is, or wants to be, a programmer, and Ning didn't get the kind of traction of a MySpace or YouTube. Nor was it likely to.

Faced with that predicament, my suggestion to Ning might have been to pitch it as a white-label development tool for third-party resellers : small agencies who are creating web-sites or intranets for other companies. Or maybe something to do work on enterprise intranets. However they clearly wanted something else.

Now what we've got looks suspiciously like Tribe given a makeover by 37 Signals.

Which is mysterious. I'm a member of Tribe, and I like it very much. But it's widely considered to be a "failure" in the YASN market. What makes Ning think that the business model is viable? Or that they can make it work?

One thing that occurs to me, maybe this is really the old Ning in disguise. Perhaps a lot of social development stuff is there behind the scenes, and this is simply wrapping it in something that looks more familiar to the general public. That would make some sense, use "make your own tribe" to draw in people who wouldn't have dared go near the old Ning, but then just happen to have a bunch of Ning-apps. sitting around which can be add to these networks. And which the more curious users could learn to customize and share.

However, having gone to play around a bit, that doesn't seem to be the case with the new Ning. Contrariwise, it looks like history has been ruthlessly rewritten; Ning was allegedly founded in October 2004 to give everyone the opportunity to create social networks. The old Ning was full of "HotOrNot" style comparison games and mashups with Google Maps. But while the new Ning allows you to add discussion forums, photos and videos to your tribe social network, hotornots and mashups are conspicuously absent.

Maybe the developers' docs will cast more light on things. You can still program stuff, but there's no hint of users sharing their code with each other, which was the really interesting part of the old system.

So why does Cisco want Tribe's code? Beats me. Speculating wildly, maybe there's something to do with convergence in the home between television, internet, entertainment etc. And maybe the next step beyond BitTorrent peer-sharing and You Tube is some kind of private P2P video sharing within social networks, so that tribes become an organizing principle for video, rivaling television channels.

But that's speculating wildly ...

Anyway. I'm a member of several YASNs but only two mean anything to me : Tribe and Linked-In. I joined Linked-In years ago, and pretty much forgot about it. It's a pretty inert sort of service. I wouldn't think about it or visit the site at all except that people I know keep joining it and linking to me. I've started to come to the conclusion that there's something rather interesting about this extremely slow social network. Most importantly, it's all about the world outside the web and outside web-time. You don't go there, don't do anything with it. But slowly and surely the network keeps accreting more ex-colleagues at the glacial pace of 4 or 5 a year.

That's all it does. But it's more succesful at that than anything else. I tried to recruit my IRL friends to Tribe. But only a couple of the more web-oriented ones bothered to join. Yet people who have no time or inclination to be on the web keep turning up on Linked-In. Whereas Tribe is great for meeting new people and having valuable conversations, Linked-In turns out to be great for reconnecting and tracking "old" people. I guess it's the combination of seriousness and undemandingness; people join it who have no time for web-culture and who are scared to become involved in the more active YASNs.

Anil Dash on Yahoo Pipes

Points out what a good IDE it is (and analogy with Ning)

The other platform I'm going to be watching is, of course, Yahoo Pipes.

By weird synchronicity, yesterday I wrote an email to someone where I was talking about ideas that intrigued me, and I was thinking of examples where you'd want a fluent dynamic gestural device like, say, a pen-mouse or a Wii controller for real-time programming.

And, looking for an example, I went into a crazy rant, inspired by Seb Paquet's Algebra of Feeds.

Here's what I wrote :

Or maybe you can create software "in real-time" by, for example, routing and mixing RSS feeds. Eg. you'd have feed-traffic-controllers, pulling together, mashing up and mixing the outputs of different web-services, maybe by dragging and dropping, or sketching pipelines with a pen or Wii-like controller. 10 years ago, Philip Greenspun needed his own server and to be a serious programmer to make the Bill Gates Wealth Clock. 4-5 years ago, smart people were mashing up Google maps with other services with little bits of glue script. Now we have reblogging services that help automate the process of getting a feed from one place and pushing it elsewhere. And Ning which is a sort of platform for creating mash-ups and reusing other people's code.

Jump forward 5 more years. You can imagine feeds of "objects" (data + encapsulated behaviour) being published. And enterprising people noticing that objects from feed 1 have input signatures of this format, and objects of feed 2 have output signatures of that format and it only needs them to be wired together in the right way, for us to have this useful combination. So how to define the wiring? Why not just draw something like a patch diagram in a virtual modular synth? A diagram of Unix-like pipes. How frequently does this need to be done? As data gets more dynamic, you might want to have people updating the wiring diagrams daily. Why make it hard work? Why not just let them draw it with a pen, or Wii controller.

And now, to quote, Tim OReilly :

Yahoo!'s new Pipes service is a milestone in the history of the internet. It's a service that generalizes the idea of the mashup, providing a drag and drop editor that allows you to connect internet data sources, process them, and redirect the output. Yahoo! describes it as "an interactive feed aggregator and manipulator" that allows you to "create feeds that are more powerful, useful and relevant." While it's still a bit rough around the edges, it has enormous promise in turning the web into a programmable environment for everyone.

These are very interesting times.