�ݺ�ߣ

Editor's Notes

• #2: Me again, for another year. I’m David Coveney. I think the committee just doesn’t learn, do they? Still, if you’ll bear with me for half an hour or so, I’m going to bang on a bit about WordPress in the Enterprise. [click to enterprise pick]
• #3: Ultimately, it’s pretty hard to get Enterprise contracts. But if you try really really hard!I did hear actually that Sulu recommended WP 155.3 for the ship’s intranet but they opted instead for Drupal 302.1Anyway, I guess you’re more interested in this kind of enterprise, right?
• #4: This type…But before we move on, we have to define what’s meant when we talk about Enterprise. It’s more about large scale, large risk projects and work, rather than big business. So it can be relatively simple work for a big business with a reputation to lose, or it can be complex work for a small business whose future depends on your work. That’s why the startship Enterprise was so named.So we ask the question - can WordPress make it in a big business like Reynholm Industries? How can you get a slice of the action? Can you implement WordPress in your Enterprise? Could you succeed at Reynholm Industries? And I am here to help explain why WordPress and YOU, can handle enterprise projects.However, there may be some disappointment in this project as, erm…
• #5: No skittles!Yeah, sorry – but your dentists rang up, thanked me for the business, and suggested I don’t do it again.And another promise… no song! This is the reaction at the office when I told them they wouldn’t have to listen to my rehearsal this year…
• #6: Anyway… [click]
• #7: I’d like to share a little story, if I may. In April 2010, the volcano Eyjafjallajökull (ey-ya-fia-la-yo-cull)…. Oh heaven knows how it’s pronounced, but it’s a volcano. In Iceland. decided to get a bit excited and it spewed a vast cloud of ash out towards Europe. It looked like this on the satellite images… [click]
• #8: It was magnificent! And caused no trouble! Unless you were flying, of course… as it turns out planes don’t like ash clouds! [click]
• #9: Many travellers in Europe faced departure screens like this. Everything cancelled. It was a mess! And that made the people at NATS, the air traffic guys for the UK, very busy indeed. Before long, their website looked like this…
• #10: 503 errors.Though I actually think 503’s should look more like this…
• #11: NATS knew that come the next big event, they wouldn’t be ready. It was time for a new platform. And look at their lovely site now…
• #12: It’s built on WordPress by Steve Sandisson, our contractor down there, and Matt Mills from NATS who deals with external communications. I think you’ll agree that it’s lovely… And that’s WordPress, being used in a powerful way, with lots of external data pulled in, by a big and serious business of 5,500 or so people.Let’s flick through a few other high profile sites running WP…
• #13: These guys had burned through a ton of cash with an implementation using a high-end CMS. We rebuilt a one year project in three months, reduced hosting costs to one fifth, got uptime up to the target levels, and gave them 30x the traffic capacity!
• #14: This one’s interesting – Plaza Cinema is an independent community cinema run as a charity. It wasn’t a difficult project, but by the use of the movie database APIs to help build their content and do their picture research we reduced the time taken to create a film listing by 80%, freeing up about ten hours a week of time for their very small team, and that meant they could do other valuable work.
• #15: We had some involvement in this until we stupidly recommended they hire Paul Gibbs and they suddenly disappeared from our accounts! Thanks, Paul! But they do a lot with WP… successful project. These sites are all ones we’ve had some involvement in. Add in Human Made, Code for the People, and you have an amazing roster of corporate, media, and government websites.
• #16: It all proves one thing… everybody repeat after me…Louder!So… We know this. Why doesn’t everybody else? Why is the perception sometimes out of kilter with the reality? Why do we even have to have this discussion? If WordPress is so ace and solves so many big problems why aren’t there more Porsches in the car park out there as enterprise projects flock to our doors? And perhaps it’s because we don’t talk enough to possible clients about what WP can be really great at, possibly because the core team aren’t interested in making the base product something immediately attractive for enterprise projects.But let’s not complain, and let’s concentrate on the good bits… here’s what WP is good at…
• #17: You can do large intranets.But before you get too excited, you’re not going to lock horns with Sharepoint. Trust me, WordPress and SharePoint are not competing in the same space even though there is some overlap. Of course, a bad SharePoint implementation is expensive hard work and that’s where it gets its mixed reaction from. And it’s not designed for publishing sophisticated websites, though for heavens sake some people insist on trying. But document management? Complex document management? It blows WP out of the water. It has the security models, the close integration to LDAP and so on. It’s really very good. I suppose you could use Lotus Notes, but… naaaah.Where WordPress *does* win is to create easy to publish departmental intranet sites, team sites, and more. It’s ace for that. You can easily gather data, run surveys, create microsites and so on. WordPress doesn’t do *everything* but it can certainly do a lot, for very little money.
• #18: Busy websites. Oh yes, WP can do that!There are so many busy, cheaply run websites that this argument is long won. Use WordPress VIP if you’re scared of servers or don’t know good people for the job, but when you can use a couple of VPSs and a CDN to pump out a million page views a day like we do for some of our clients, you know it’s pretty good. Need more speed? Buy more servers. No problem. <think about language here – broader audience… talk about costs… eg, £500 a month would increase capacity in a day or two>
• #19: Big websites – how big do you want? Ten thousand posts? Fifty thousand? No problem.All big websites, regardless of performance, will hit performance and management problems if the structure isn’t carefully planned. Computers still aren’t powerful enough to make up for stupid decisions. So a bad permalink structure or some daft queries and your site will grind.
• #20: A lot of large organisations need to quickly create new microsites for specific needs. One client of ours duplicates and customises similar sites. In the past they spent £1k per site on a web developer. Now the dev cost per site is approximately £50. Not bad eh? Yeah, we might be making ourselves unemployed here…
• #21: You can do a lot with custom post types and terms. A heck of a lot. And you can present a fair amount of data to a fair amount of people. But you can’t use that method to present a huge and fully searchable database of millions or billions of rows. Try to run Rightmove or similar on WP without custom tables and it’ll reach a point where it all gets too much and falls apart. Brilliant, however, for proof of concepts. Repeat after me – WordPress is not a fully functional Relational Database Management System with extensive and scalable indexing capabilities [nobody will – laugh say “What? Oh OK.”]
• #22: So… this brings me on to the next part of my talk. How can we make WordPress better for clients with big needs, and so make our marketing and sales process easier?
• #23: I really really think WP should have robust caching and CDN support built-in – it’s essential. I suspect one reason it’s not there is to help make VIP services more attractive. If you understand caching and a cdn it’s almost trivially easy to set up a site capable of handling 2000 concurrent users on a basic vps costing you £50 a month. Without caching and a CDN, this is the sort of response you get from your server the minute a post gets any interest on a subreddit, let alone the front page.
• #25: It’s all a little clunky right now – it needs groups and better multisite handling. Imagine you have ten websites, five need to be managed by team A, five by team B, and you need to three editors that oversee both sites and approving content you basically have to set up each user on each site. Imagine you have fifty sites and two members of staff change. That’s a hundred edits you have to do. Slow slowslow. And then the security handling is so daft…
• #26: This a client’s typical reaction when I tell them there’s no safe way to give Editor’s access to widgets without blowing open other admin options. You can /hide/ things of course, but hiding stuff doesn’t make it inaccessible to the wily.
• #27: If you don’t get workflow right then at times, during their working day, a client might feel like this as they try to get something done but find they can’t… But without good user and blog grouping, workflows across sites become very tricky.
• #28: This is how a new employee at a large business with multiple blogs on multisites ends up feeling because setting them up was so tricky they only got added to the ones the busy manager remembered to do.
• #29: This is the kind of reaction I get from UK based clients when I tell them the cost of VIP services.
• #30: Automattic’s VIP services appear to be great – but a starting point of $3,750 US a month is a bit rich for most UK websites whose market is substantially smaller than that available to a US provider – consequently there are relatively few UK VIP hosting clients. But the only way to get access to a lot of information, support, certain plugins and guidance for WP is to use VIP services or VIP partners. I don’t really mind that, and we’ll probably go in that direction soon, but I think if WP is to have more take up the VIP network needs to grow beyond the two current UK members.
• #31: I’m not going to say much on this, but how many of us think “yippee!”every time a patch or a new version comes? I’m more like this…[click]
• #32: Every it seems, day, something needs to be upgraded, and it’s almost impossible to tell if you *have* to upgrade or not. This is how I feel when I get in and find yet another plugin has been shown to have a gaping security hole that’s been published on every hacker list.
• #33: And maybe it’s time to get serious…The whimsy evident in some parts of WordPress just doesn’t work well when you’re dealing with very serious people. It’s not *awful* incidentally, but “howdy $username” isn’t great. Hello Dolly needs to die. But there’s worse - one very big client of ours had a very very worried user that needed a lot of calming down when they saw this…
• #34: Hands up who recognises this?
• #35: Functionally and technically, WP probably *is* ready for broader acceptance in the enterprise. But culturally it isn’t. We need to do things to make it a better fit…Well done, WordPress, well done for showing something saying “Self destruct” to our clients… Please note, this is now gone from the forthcoming WordPress 3.6!
• #36: We wrote The Auditor, a tool for logging settings changes, content updates, logins and so on, in order to show clients that they weren’t being hacked by cleverness, they were just very very bad at passwords. When you show someone a graph like this…
• #37: They’re suddenly pretty easy to sell stronger security to. It’s both an auditing tool *and* a sales tool, and available from us for just $250. Just a little plug there, but my real point is that auditing in WordPress would be so much better if it was plumbed into core as there aren’t always the right hooks available and getting things perfect is really tricky. And enterprises love proper audit trails – it covers people’s arses.So this brings us a problem…
• #38: Some of these changes require substantial work, or the integration of plugins in a way that adds functionality to WordPress that simply isn’t wanted by the core developers who mostly work for Automattic, Aubrey Capital or the WordPress Foundation – all controlled by Matt Mullenweg – a charming, nice guy, but also a human who is naturally going to look after his best interests. As WordPress.org and WordPress.com won’t diverge much, there’s not much hope of this stuff hitting core, and if Automattic *are* into these things you’ll only see it through jetpack and VIP exclusives – both of which have their issues for many of us wanting to get the big contracts in smaller economies than the US.Does this mean a fork, or a distribution? Do you know what I think the community will do to whoever forks?
• #39: Well, maybe not – but a fork could be seen negatively by the community. More likely, I think, if you want to stay in the WP fold, is going to be a distribution with packaged plugins, like the standard WordPress release has Akismet and a theme. It would be quite a project in its own right, but it would be a good starting point.So can we create an enterprise friendlier version of WP? Well yes… yes we can… - I don’t know if it’ll come through marketing a product based on standard WP and a set of plugins, or a fork, or what… but win enterprise…[click]
• #40: Yep… we’re going to be happy… it’s a big big market trying to solve big problems. Big businesses have big reputations to protect and a lot of things to promote. If you can make a business more efficient, or better at what they do… they’re interested. Trust me.
• #41: So… last part. What are the benefits to WordPress?
• #42: We all know this – no massive license. You can set up a site for thousands of page views a day, that looks good and markets a product in a day and not spend a penny on outside suppliers. Amazing. Not only that, but many proprietary systems are well into six figures before you’ve even started on customisation… and then there’s the hosting fees… we’ve come across some shockers.
• #43: And that means lots and lots of people with WordPress skills are available. If your supplier turns out to be awful, or your new developer is lazy… get another one! It’s not hard.
• #44: What seems like WP’s weakness is also it’s strength – its very simplicity makes it more of a fundamental tool than something that tries to be all things to all people. You can add what you need to make it do what you need. And simplicity means you can pick it up quickly and easily.
• #45: If you need to move quickly in your company, WordPress can be your friend. Don’t stress it – get on with it. Ship and make money.
• #46: If somebody has a business model for some free software then that means it can continue and can have a long life. You just know WordPress will be around for at least another ten years – it’s the winning platform, it’s popular, and the company that pays for most of its development is making money. That is very reassuring for businesses. No abandonware here.
• #47: We make sod all money from training our clients, dammit! Because really, if you’re halfway smart, you can work WordPress. That means our training business plods along, but we won’t be rich on it!
• #48: According to the digital editor at The Spectator, since switching to WordPress they’ve more than doubled their traffic. No big SEO drive, no PPC campaigns, just good content and a good CMS. WordPress has been a big part of that. Many of our other clients have seen similar growth in their digital properties… and it’s great to see. We keep clients because they’re happy. They’re happy because of us, but also because of the CMS and excellent cost control.
• #49: One thing I’ve been really impressed by in WordPress land is the people. You’re amazing. I’m looking forward to the next ten years of WordPress. So I’d like to clap my hands to you, all the various contributors, plugin authors, theme designers and business men and women who make WordPress a success. [click]
• #50: [wait a sec for loop… then click]
• #51: If you want to afford a mortgage on a house in today’s crazy world – go out there, make convincing cases, and be successful implementing WordPress in enterprises. Thank you!