I do not understand why successful technology companies, seeing another different but also successful technology, realize "we can do that too" and immediately assume "and therefore we should".
Microsoft sees Google being a search engine and sinks billions upon billions into failing to be one. Google sees Facebook being social and decides it needs to be social (an upcoming disaster). Facebook sees Twitter being real-time and decides it needs a real-time stream too (a bad idea they have since buried under the "most recent" link).
In the CMS world, we have Drupal for complicated content-management, Wordpress for full-featured blogs, and Tumblr for drop-dead-simple blogs. There is no reason one of these products should try to fill all three roles, and no practical way to do so (because it is, I posit, impossible to be both as simple as Tumblr and as configurable as Drupal).
Wordpress is great. It is the leader in its category. There's no need for it to try jumping into another category that would distract from its core competency.
I do not understand why successful technology companies, seeing another different but also successful technology, realize "we can do that too" and immediately assume "and therefore we should".
Is it really that hard to understand? Most corporations, almost by definition, must grow...often in multiple directions. They must expand market share in markets they already exist in and enter new markets they previously shrugged off.
I wouldn't say companies do this "immediately." In fact, quite the opposite. Microsoft joked about the search market for years before realizing its more than a joke. Ditto for google and social networking...social was seen as a joke at google circa 2006 as facebook was on the verge of exploding.
>social was seen as a joke at google circa 2006 as facebook was on the verge of exploding
"In 2003, Google offered to purchase the social network Friendster, but the offer was declined by that company. Google then internally commissioned Orkut Büyükkökten to work on a competing independent project. The result was Orkut. The product launched on January 24, 2004."
A failed purchase of a social network only proves my point, if anything. When google really believes in a company, they go to great extremes to grab it. See YouTube.
This was 2003. Facebook didn't even exist then. Are you just trolling? Would a big YouTube-style purchase of Friendster have even worked out well for Google?
I understand expanding into adjacencies as a sensible mode of growth for a company with mature products -- so Google obviously expands into different types of advertising, and AOL into different types of content, and Apple into new form factors of portable electronic devices.
But I think the critical difference is that those are new products by old companies, and specifically products where they can use existing expertise or technology and apply it successfully. So Microsoft's expansion into search makes no sense because they hate the web and have always sucked at it. Facebook's expansion into real-time made no sense because the whole value of their stream was its curated nature.
Of all of the examples of bad expansions I mentioned, the suggestion that the contributors of WordPress create a "WordPress lite" that is simple and Tumblr-like actually sounds on the face of it like the best of the bunch. But in my opinion, Tumblr's primary advantage is not its interface (though that's a big secondary feature) but its community. Tumblr provides a built-in mechanism for content discovery and distribution, and is insanely viral -- the only way to comment on a post most of the time is to reblog it to your own tumblr.
For WordPress to create a Tumblr-like "WordPress Lite" would involve far more than merely stripping down the interface. They would have to completely rework WordPress.com to provide a centralized identity and content-distribution mechanism. Not only would that take them far away from their core competency of building single-instance software, but as Tumblr is demonstrating, building such a platform is technically very challenging.
My point is not so much that WordPress should not attempt such a project. They're smart -- maybe they could pull it off. But such an expansion is by no means a simplification of their current software; it is in fact a significantly harder problem.
Is it really that hard to understand? Most corporations, almost by definition, must grow...often in multiple directions. They must expand market share in markets they already exist in and enter new markets they previously shrugged off.
Why "must" they grow? Where is the line drawn for "most" corporations? I have a feeling this sentiment is simply parroted as received wisdom.
It is an article of faith in contemporary business that a company not integrating vertically or horizontally is doomed. The Peter Principle, the tail wagging the dog, take your pick: the lesson has been learned everywhere but the board room.
...which is why I've found it weird that people try to shoehorn WordPress to become a full-blown CMS platform. There are better tools for that, and the effort detracts from the strong points of WordPress in blogging
Having looked at the field numerous times when building and advising on websites for smallish newspapers and magazines, I have yet to actually find a better CMS platform for the common use case of "I want to publish articles, plus some other additional stuff on the side," since that matches the actual structure of a WordPress-based CMS.
Even though Microsoft made its mistakes in search, I do not think Bing is anywhere near a failure of search engine. Look at Cuil for a true example of a failure.
^^ Agreesies w/citricsquid. (Due to the fact that it's highly customizable, Tumblr gets compared to MySpace. But paradoxically, due to the fact that many of its early adopters are wburg hipsters w/great taste, it also has a rep for having excellent design biases. Checkout the default themes. Gorgeous.)
But it truly is more like Twitter (or even Facebook) than MySpace due to the fact that it's for publishing (or sharing) content more than pimping a profile.
This is very hard to see from "the outside" (where people are more likely to compare it to Wordpress because it's "just a blog"). But what you don't know until you use it, is that more than half the party is inside the system. The Tumblr themes are never even seen because you're experiencing your friends + followers through a Dashboard view. You're Liking, Reblogging (not too unlike retweeting), adding commentary, asking questions (that are hidden on the outside), squeeing over famous followers, etc.
Tumblr, like Twitter is a place. You can interact with people w/out asking them to visit your-site-somewhere-else. Unlike Twitter, is has the advantage of also being a destination. That link you're sharing? That essay you wrote? It's all right here, still on the Tumblr.
[edit: Businesses see its popularity. Tumblr has as many users as Twitter did two years ago. IBM, The Economist, The Today Show (!?), most Conde Nast magazines, YC start-ups like Convore, high-end hotels... I've seen 'em all on Tumblr.]
As for what the original post is about: It's a fair question. Tumblr is easier than WP. But that's not why it's popular. It's the social part. While we wait for our utopian federated, distributed social networks to arrive, these centralized places are pretty great.
This is a hugely overlooked point. There's a huge variety of simple blogging tools, but Tumblr's difference is the simplicity combined with the community.
The ability to reblog with simplicity is a huge part of their success.
I just moved the SproutRobot blog from Blogger to Tumblr for one reason: People share stuff on Tumblr. No one really shares stuff on blogs.
You can add a "tweet this" or "like this" button to your blog posts, but the only way to get Tumblr kids passing your stuff around is to have a tumblr. Moving away from Blogger (or Wordpress) doesn't seem to sacrifice any shareability, which is the singular purpose of blogging.
I've had blogs on LiveJournal, Wordpress, Blogger, Posterous and Tumblr, and here's my view on the difference:
Livejournal: The proto-Facebook, for personal blogging and sharing your thoughts/lives with your friends, with a personal connection with the people who follow you.
Wordpress.com/Blogger: Mainstream blogging, letting you share your opinion with the world. I switched from Wordpress.com to Blogger because they inject ads into your blogs. Neither have great editing interfaces.
Posterous: Mainstream blogging simplified, I switched from Blogger to Posterous because it just works. I can write a post insert images and it will come out looking decent (on Blogger/Wordpress I have to spend far longer tweaking layout, etc.).
Tumblr: Single media micro-blogging, Tumblrs support for mixed media blogging is very poor, you have a photo or a block of text, but not both. It's much more about community as well, it's closer to Twitter than to the other blogging platforms. It's all about followers and re-blogging other peoples posts.
I don't think it's Tumblr, but what about Posterous?
For websites where the blog is just a part of the site and not the central component, Wordpress is overkill. And maintenance of updates and plugin updates is a pain once you start getting beyond one or two installs.
I've been thinking that it's time to reevaluate the blogging options out there rather than defaulting to a big-ass Wordpress install for every client that just wants a simple blog as part of their site.
Postereus is pretty much Tumblr for 20-somethings. From what I've seen most of the people who have blogs in Tumblr are the generation prior to Postereus.
I would make the argument (as a 20-something) that there are far more 20-somethings on Tumblr than there will ever be on Posterous.
Posterous has always felt, in its design, to be minimal and content based. It is not about posting cool photos you found online, funny youtube clips and a quote out of the book you're reading.
Has anyone ever looked at the internals of WP? You'll find all of the worst practices known in PHP development (which I find to be a bad practice all it's own). Maybe it's bias, but I can't imagine people who write such poor code being able to compete with a startup that actually uses technology competently (which may or may not be Tumblr).
In this case the question is more about whether Tumblr serves a specific need better than WP and not apparently about technical superiority. Still, again, if you can infer anything about the minds behind WP from the code they write then these guys are creating an app that was cutting-edge innovative maybe ten years ago. For God's sake they don't even have a proper separation of logic from presentation.
Though I mostly agree, I sincerely hate this, and this isn't the first time I've seen it:
> but I can't imagine people who write such poor code being able to compete with a startup that actually uses technology competently
It's a pretty big jump from coding quality to inability to run a startup. You're assuming it's the same people, for one, and history has shown us that code quality has nothing to do with startup success. In this specific case, doesn't Wordpress contribute useful high-scalability software back to the community? I'm drawing a blank but I seem to remember something like that.
> It's a pretty big jump from coding quality to inability to run a startup.
Would you say that the general cluelessness of the people at MySpace was totally coincidental to the fact that they were probably the largest user of ColdFusion?
You mised my point: it was not that code quality causes startup fail but that poor technology choices are symptomatic or correlated with fail. For one, it impairs their ability to change the focus of their product.
Take, for example, the ability of their platform to support user-specified data models instead of simply posts, pages and comments. This was an ability that many users were requesting or implementing themselves (or working around) for many versions. It was only implemented in version 3.0, not even a year ago.
This is something that is trivial to support on any system designed around proper MVC principles. A well-designed competitor could have had this years ago.
From all this and more, I conclude that WP is vulnerable to disruption by a technologically superior product.
I sense a lot of idealism in you. MySpace was a Microsoft platform in the end and dropped Coldfusion, and there were far more variables related to its failure than its technology stack (contrary to what Scoble would like you to believe). I work for a company that uses Coldfusion extensively, and we succeed just fine, so your grand painting of Coldfusion as a bad decision is flawed with my sample size of 1.
> Take, for example, the ability of their platform to support user-specified data models instead of simply posts, pages and comments. This was an ability that many users were requesting or implementing themselves (or working around) for many versions. It was only implemented in version 3.0, not even a year ago.
Yeah, I'd rush to commit development to that too. It sounds pretty useless except for people who are molding Wordpress into something it wasn't envisioned to be.
Far be it from me to defend Wordpress's code quality, but they have two things going for them: (1) word-of-mouth and (2) install base. They got there using the code base they had, and it didn't get in their way. With a large install base comes responsibility. Rewriting your system and disrupting a significant plugin install base comes with a list of cons that grows and grows.
Proper code is a misnomer. There is no such thing as proper. Also, MVC doesn't apply to all use cases, and it certainly isn't a prerequisite to "proper" or "well-designed" as you imply (even when MVC is a perfect fit, there are alternative approaches). Wordpress went to market with what they had and they dominate the market.
> From all this and more, I conclude that WP is vulnerable to disruption by a technologically superior product.
Technologically superior products exist. Wordpress continues to dominate. Why is that, you think?
First, this argument is about the successfulness of the WordPress internal architecture. If you haven't looked at the internals WordPress, or written something against their PHP API, then you're arguing from ignorance.
> I sense a lot of idealism in you.
Please do your best to limit yourself to relevant statements.
> I work for a company that uses Coldfusion extensively, and we succeed just fine...
This is just intellectual dishonesty: you don't reveal at all whether or not the core of your business depends on Coldfusion, which would be important in order to make an apples-to-apples comparison.
You work for Linode, a company that provides VPS hosting. If you make a poor choice of technology on which to base your website, the core of your business is not impacted. Maybe some customers gripe about your dashboard, but your competitors will compete mainly on the superiority of their servers. If you work for a company like MySpace (which depends on its website entirely) and Facebook enters your market, you will not be able to find enough talented CF programmers to fend them off. Talented programmers would not ordinarily elect to use CF, for many reasons.
> It sounds pretty useless except for people who are molding Wordpress into something it wasn't envisioned to be.
Now you're simply being a contrarian. I don't think you even understand what this feature is or how it works. And yet again, I have to remind you of the point was: a feature that they wanted took a lot of time and effort to implement, when it could have been had trivially with a better architecture. In other words the technology decisions created inflexibility when implementing a product feature.
> Proper code is a misnomer. There is no such thing as proper.
First, your college professor might have an objection here.
Second, I said "proper MVC principles", as in, had they properly implemented MVC principles. I would be happy if they implemented any sort of architecture. This is one area where, if you had knowledge of the internals of WordPress, we could have a productive conversation, instead of simply negating what I said in the abstract.
> Also, MVC doesn't apply to all use cases, and it certainly isn't a prerequisite to "proper" or "well-designed" as you imply...
Not implied at all. Had you any knowledge of the internals of the WordPress architecture, it would again be useful to assert that knowledge so we could discuss whether or not it conforms to any notion of design. Unfortunately, you don't seem to want to discuss the actual point of the argument, preferring instead to deal with (ironically) ideal notions.
> Wordpress continues to dominate. Why is that, you think?
Past performance is not indicative of future results.
This kind of argument can be constructed for every company which currently leads its market, but may in the future be overtaken by a competitor. The relevant question is not "why does it dominate now" but "is it threatened by X".
I have to disagree with you here. Just because someone writes poor code doesn't necessarily mean that they have the business sense of a rock. If someone writes crappy code but can market it better than the other guys, they'll probably win. I think Tumblr and WordPress both excel at blogging depending on the end user's needs. While they both essentially do the same things, each has its own merits. People who are looking to get into blogging or networking of any sort should definitely consider both. If they've only heard of one or the other and decide to go with it without researching, whatever they choose will probably work best for them.
Having waded through WordPress's code base for a few years, I agree with you to some extent. However, most people aren't going to care. Do you think half of the people who use MySpace, Tumblr, Twitter, etc care what it's written in or what it looks like inside? Facebook and Yahoo even use PHP and seem to do alright (though I admit they've modified it and have money to throw at problems to get them fixed). While I'm sure WordPress's code base drives some people away, people who don't know any better simply won't care.
I feel like I have pretty good insight on this issue.
I actually use both. WordPress is my backend for Tumblr. I run a news blog called ShortFormBlog ( http://shortformblog.tumblr.com/ ), which has content that works well on Tumblr but requires a more robust approach in terms of the creation of the content due to the fact that the posts tend to be more graphical in nature. Which is where WordPress comes in.
It has a lot of benefits. When breaking news happens, I don't have to cover every piece of the story … I can reblog the missing pieces from other news Tumblrs. And if I want to do more detailed posts, I can switch back to my customized WordPress backend, which has a modified version of TinyMCE and Tumblrize to allow me to do more design-heavy posts. I also get to keep an archive of the stuff that took more time so it's held in a non-centralized place.
The overall benefit? At 6,000 followers and counting in just six months, I have a much larger audience than I ever did with the WordPress site. There's much more communication with readers, and since it's centralized as opposed to all over the place, it's easier to latch onto. I also have gotten some media coverage from Mashable and Mediaite, and I've even worked with the Tumblr staff on some things.
WordPress has its uses, but Tumblr has major advantages that you can't get with a self-hosted site. The community is there and it's been well-nurtured. That's as good a reason as any to give it your attention.
I have never known a time where Wordpress was not way more bloated than Tumblr is now, and I've been using Wordpress for about 5 years. That being said, the comparison is only valid when comparing them as services - as far as I know, Tumblr is not available as a platform to use on your own server. (That sucks, because the Tumblr service is extremely unreliable.) As services, though, Tumblr is way more fun to write on.
>Putting aside the flack Tumblr has gotten for their downtime issues of late,
I love how the author just brushes this off like it's not a big deal. I've tried to use tumblr because everyone goes silly for it, but I wasn't able to even set up an account because the site was down. That's fine for something like twitter but if using tumblr as a publishing tool for my business, what am I supposed to do when it's down?
Tumblr? No thanks. I'll take a wordpress site on my own servers all day long. If I really want some other platform to simply publish on, posterous is drop dead simple and doesn't go down.
in an ideal codebase, such as might result from vertical-slice features being added iteratively plus thoughtfully designed features, feature groups and permissions, could a mature product like wordpress not present different levels of simplicity? "enable simple blogging only" v "enable full cms"
I have first hand experience with this and I think it is not only possible, but a reasonable goal for any well designed product. (the advantage of dynamic feature walls is already a huge win)
Wordpress, like Microsoft, are bound by the golden anchor of backwards-compatability. One of their big selling points -- the multitude of themes and plugins -- is also what prevents them from making any radical architectural improvements.
They're dumping PHP4 and MySQL4 support in WordPress 3.2. You're right, they have to consider support for existing themes and plugins, but they're not chained to the past.
They're going to start using PHP 5 features, but that won't remove the large amount of PHP 4 code they already have. Neither will it remove the vast body of plugins/themes from play, most of which are updated quite sporadically.
If there's anything that will cause internal architectural changes, it's the project's general focus on new features over any other goal.
I think it is, it's really easy to create a theme or port a static site over to make it dynamic. Half of tumblr's users are teen girls but they are geeks so they make lot's available for us geeks to use it to hack.
If only Tumblr would work well, I would love to switch, and move all my family to Tumblr, too. But somehow the fancy interface never works when I try it.
Microsoft sees Google being a search engine and sinks billions upon billions into failing to be one. Google sees Facebook being social and decides it needs to be social (an upcoming disaster). Facebook sees Twitter being real-time and decides it needs a real-time stream too (a bad idea they have since buried under the "most recent" link).
In the CMS world, we have Drupal for complicated content-management, Wordpress for full-featured blogs, and Tumblr for drop-dead-simple blogs. There is no reason one of these products should try to fill all three roles, and no practical way to do so (because it is, I posit, impossible to be both as simple as Tumblr and as configurable as Drupal).
Wordpress is great. It is the leader in its category. There's no need for it to try jumping into another category that would distract from its core competency.