Tag: Google (page 1 of 5)

HOWTO: Ditch Gmail for self-hosted webmail

Update (January 2015): Although the self-hosted approach detailed in this post worked really well for the six months I tried it, I’m now using Fastmail.


On Friday, I decided to ditch Gmail. It had been a long time coming, to be honest.

I’m not really interested in having a debate whether or not I ‘should’ do this, or whether it’s objectively a good thing to do. I just had an epiphany when I realised that almost all of my data (e.g. search, email, analytics) was going via Google’s servers. It’s like some kind of legitimised man-in-the-middle attack.

Instead of Gmail, I’m using webmail on my own domain and (shared) server. It was a pretty straightforward process. Here’s how to do it:

  1. Set up self-hosted email inbox
  2. Forward (and archive) email
  3. Import folders and email
  4. Update email addresses around the web

1. Set up self-hosted email inbox

I’m using Reclaim Hosting, which comes with something called ‘CPanel’ installed. This makes it much easier to install and maintain apps and services.

Sign up for a web host that has CPanel. Login, and go to the ‘Mail’ section of CPanel:


Click on the ‘Email Accounts’ option. Fill in the email address and password – for example, I chose mail@nulldougbelshaw.com


You can now access your new email inbox by appending ‘webmail’ to your domain name – e.g. dougbelshaw.com/webmail. You have a choice of interfaces to access your email inbox. I chose Roundcube:


2. Forward (and archive) email

The first thing you want to do is make sure that you continue to get the email sent to you in one place. To do that, you need to forward all of the email that comes to your Gmail account to your new self-hosted email inbox.

In Gmail, click on the gear icon and go to ‘Settings’. Once there click on ‘Forwarding and POP/IMAP’:


Choose ‘Forward a copy of incoming mail to…’ and input the self-hosted email you set up in Step 1. You’ll have to confirm that account by clicking on a link that Google send to your new email address.

I chose to ‘archive Gmail’s copy’. It’s up to you what option you choose here.

3. Import folders and email

This is the bit that takes the longest. In fact, you can leave this running overnight and/or be doing Step 4 while this is happening.

Unless you’re declaring email bankruptcy, you’ll need to transfer your existing emails and folders from Gmail to your new self-hosted email account. Step 2 only redirects all new emails received.

First, you’ll need to use an email client like Mozilla Thunderbird (cross-platform) to download all of your Gmail folders and emails. To set it up, download Thunderbird and then launch it.

You’ll see something like this:


You’ve already got an email address, so click ‘Skip this and use my existing email’.


Add your Gmail account first by entering your name, email address and password. Remember, if you’ve got two-factor authentication turned on for your Google account, you’ll need an ‘App password’ from your security page.

The default option is to connect via ‘IMAP’ which is what we want so leave it as it is and press ‘Done’. You should then see Thunderbird importing all of your folders and emails. This will take a long time.


Note: ‘folders’ in Gmail are known as ‘labels’.

Now you need to add the self-hosted email account you set up in Step 1. To do this go to the menu and choose ‘New’ and then ‘Existing Mail Account’:


You’ll see the same box from when you added your GMail account. This time add your email address and password to the account you set up in Step 1.*


Click ‘Done’ and you now have both your Gmail and self-hosted email account syncing with Thunderbird.

Now you need to select all of your Gmail folders/labels and drag them to your self-hosted account.


Be careful to drag them onto the email address rather than ‘Inbox’ – otherwise the folders you drag will become sub-folders of your inbox rather than folders in their own right. Of course, you can always just drag them to ‘Archive’ if you don’t care.

Now wait. Possibly a very long time if you’re on a slow connection and/or you have lots and lots of emails and folders.

4. Update email addresses around the web

While your email and folders are transferring – and, to be honest, over the next few days/weeks – you’ll need to update your email address with the accounts you have around the web. There’s no great hurry for this, as your Gmail messages will be redirected to your new email inbox, but it’s nice to get things sorted.

You may also want to do one or more of the following:

  • send a ‘please update your addressbook!’ email to your contacts
  • use an email auto-responder on your Gmail account for a while
  • add a message about having a new address to your email signature

EDIT: Remember, if you don’t tell people about your new email address, your emails will still be going via Google’s servers (thus negating the point of the exercise…)


Did you manage to follow these instructions? Have you got a different/better way of doing it? I’d love to hear from you in the comments section below (or via this Hacker News thread)!

*Apart from believing in Open Source software (and working for Mozilla), one of my reasons for using Thunderbird is that it provides auto-setup for a much wider range of services than other mail apps. Also, the reason you see ‘SSL’ here is because I set up https on my domain using StartSSL. That’s outside the scope of this tutorial, but is also probably unnecessary if you’re planning to access your inbox via the webmail interface.

Banner image CC BY David Huang

Software with shareholders (or, the menace of private public spaces)

TL;DR: we’re using software with shareholders and interacting in private public spaces. We can do better than this.


I live in Morpeth, Northumberland, a lovely market town in the north-east of England. It’s the kind of place that still has a vibrant high street and plenty of stuff going on. Somehow or other it’s survived the hollowing-out of places that seems to have accelerated since the 2008 economic crisis.

Offline private public spaces

Within Morpeth is a small shopping centre called Sanderson Arcade. It’s got shops like Marks & Spencer and Laura Ashley, piped music and a friendly, retro vibe. To help with this they employ men called Beadles, ostensibly to welcome people, give directions and lend an Edwardian air to the place. If you pay attention, however, the Beadles wear earpieces and both look and act a bit like bouncers. It’s then you begin to realise that one of the main reasons that they’re there is to keep out the riff-raff.

I’ve got no particular problem with Sanderson Arcade nor with the Beadles. I would be interested to see what would happen if a group wanted to stage a protest there. I guess they’d invoke the fact that it’s a private space pretty quickly. Still, there’s other places in Morpeth you could go to protest and still be seen and heard. Sanderson Arcade isn’t the only place people go, and the other spaces are owned communally. That’s what our taxes are for.

Online private public spaces

The problem comes when we apply what I’ve just described as a lens through which to understand what we do online. I suppose that, yes, we’re surveilled on CCTV within offline public private spaces. People can track what we purchase. We can answer survey questions about our shopping habits and lifestyle. But that’s where we hit the limits of the analogy. Online private public spaces are very different.

I came across this today. It’s the latest in a long list of examples demonstrating the amount of data Facebook collects on its users. And it’s not as if everyone is unaware that Facebook, at its very core, is a scary privacy-loathing service seeking to track as much about you as possible.  Once it has that information it sells it to the highest bidder. I like to think Google’s slightly better in this regard, but if I’m honest that’s only because I use Google’s services more than Facebook’s.

Almost every space in which we interact with other people online is a private public space. For me, Twitter and Google+ are prime examples. In the past we’ve been reassured by Google’s mantra of “don’t be evil” and how people in Iran and Egypt used Twitter to rise up against their oppressors. The reality is that both of these companies are now companies in which you can buy stock. They need to deliver shareholder value.

The problem

I’m increasingly leaning away from using software that has shareholders and leaning towards alternatives. I’m writing this on a Chromebook with Ubuntu 14.04 installed, for example. One of the things that’s great about non-profits making software is that they can innovate on behalf of users, rather than in ways that will increase market capitalization.

The problem we’ve got is that to interact with other people you need a means of communicating with them. When everyone’s physically co-located you can use your voice. If the place you’re in is unwelcoming or not to your liking then you can all decamp and move elsewhere. This is not the case when you’ve got a network of thousands of people distributed around the world. It’s quite likely that the only means you’ve got of contacting one another is through a single privately-owned platform.

I’ve toyed with the idea of closing my Twitter and Google+ accounts many times over the past few years. The problem with that is that it would affect me professionally. Not only are they spaces from which I gather a lot of information to do my job, but they’re spaces where other people find out about my work. You can’t do good in the world as a knowledge worker if no-one knows what you’ve been doing.

Our response

So what can we do? It’s not a problem for us to solve as individuals, but something for us to do collectively. And the call to action can’t be protect your privacy! because, to be quite honest, people don’t seem to care. Technically-minded people think that building a version of Twitter or Facebook or WhatsApp but with public-key encryption will see users flocking to their site. Well, here’s a newsflash for you: no they won’t. They’ll trade privacy for convenience.

Instead, we need to work at a meta level and do some systems thinking. Here’s a bad idea: try to get everyone to switch from Twitter to IRC. Here’s a good idea: work on creating a compelling way for users to bring their own data and authentication to services. Unhosted seems to be on the right track with that. Just because things have failed previously (OpenID!) doesn’t mean they’ll necessarily fail again.

Of course, the meta meta level problem centres around online business models that aren’t dependent upon advertising. Providing free services and selling user data is the high-fructose corn syrup of the internet. While we in the west might tolerate paying for services, many of those getting online for the first time in developing countries know nothing but Facebook Zero.


Things change. The tech world used to be full of people resisting The Man; the discourse was around connecting people and envisaging new possibilities. Now, however, we have a tech elite with control of the spaces in which we interact. If you don’t understand the potential implications of this, then you might want to dig a little deeper into the NSA revelations and read Dave Eggers’ The Circle. That will open your eyes.

Public spaces should be public and commonly-owned. Perhaps it’s time for governments to stop fawning over billionaires with technical skills and start providing services for all of us. Maybe instead of dismantling the state to allow for private profit, we can use technology to create a more egalitarian and just society. And could we, just for once, use technology in ways other than shoring up the privilege of the one per cent?

Image CC BY Nathan O’Nions

Open Education and the Open Web (#openeducationwk)

This week is Open Education Week 2014:

Open Education Week is a series of events to increase awareness of open education movement. The third annual Open Education Week takes place from March 10-15, both online and offline around the world. Through the events and resources, we hope to reach out to more people to demonstrate what kind of opportunities open education has created and what we have to look forward to.

Mozilla is playing a role, through a week-long online discussion entitled Open Education and the Open Web. There’ll be a new question to prompt conversation each day in our Google+ Webmaker community.

What does it mean to participate on the open web? How can we encourage others to take agency over the opportunities the open web provides? This discussion led by Mozilla’s Doug Belshaw will explore the participatory culture of the web, why it matters, and what we can do to protect and cultivate it.

Today’s prompt is simple. We’re just asking people to introduce themselves and respond as to what ‘open education’ looks like in their context.

You should join us. It’s totally fine to dip in and dip out. Take the first step:

Click here to join the Mozilla Webmaker Google+ community

Image CC BY mozillaeu

More on the web as the platform

Earlier this year I wrote about my attempts to move to a web-based workflow. I discussed how people tend to see devices that rely entirely on the web as for other people.

Last week I moved house. We don’t get broadband until tomorrow so I’ve had to go about my web-based job by hopping between dongles and tethered devices. On top of this, I decided to run an experiment. Rather than using my MacBook Pro as usual, I opted to use a Chromebox connected to a 24-inch display, wireless keyboard and mouse.

So instead of picking up my MacBook Pro every time I ran into an issue, I decided to use the difficulty:

  • I bought a webcam that doesn’t need drivers and works with the Chromebox
  • When I needed to use Vidyo or Skype to talk to my colleagues I used the Android app
  • I used the opportunity to try appear.in (a WebRTC app)

What surprised me was just how easy it all was. No need to update apps. No perceptible slowdowns. No spinning beachball of death. Everything I needed to do as part of my current job was possible by using the web.

Image CC BY-NC-SA Evan Leeson

The web is the platform (or, the perils of esoteric setups)

At Mozilla we say that “the web is the platform”. It’s almost like a mantra. By that we mean that, as the world’s largest public resource, the web is big enough, fast enough, and open enough for everyone to use on a full-time basis.

To prove this, we made FirefoxOS, a mobile operating system comprised entirely of web-native technologies. But FirefoxOS devices aren’t the only ones that lean heavily on the web for their functionality. Google Chromebooks have a stripped-down version of Linux that boots directly into Google’s Chrome web browser.

The meme over the last few years seems to have been that Chromebooks (and by extension, I guess, FirefoxOS devices) are for other people – you know, the type that “just do a little bit of web browsing here and there.” They’re not for us power users.

Here, for example, is Andrew Cunningham from Ars Technica talking about covering CES 2014 on a Chromebook:

Even if you can do everything you need to be able to do on a Chromebook, switching from any operating system to any other operating system is going to cause some friction. I use OS X to get most of my work done because it’s got a bunch of built-in features and applications that I like. I use Full Screen Mode to keep my laptop’s display organized and uncluttered. I like Limechat because it’s got a bunch of preferences and settings that lets me change the way it looks and works. I like Messages because it lets me connect to our XMPP server and Google Talk and iMessage, all within one client.

That’s what bothers me the most about Chrome OS. It’s not that you can’t do a lot with a Chromebook. It’s not even about getting used to different tools. It’s just that the operating system works so differently from established desktop operating systems that you’ll have to alter many of your normal workflows. No one’s saying it’s impossible to do, but for people used to something else it can be a laborious process.

Don’t get me wrong: there’s nothing wrong with native apps. I really like Scrivener, Notational Velocity, and others. But unless you’ve got unusual requirements I reckon that in 2014 you should have a workflow that can use the web as the platform. In other words, being away from your own machine and ‘perfect setup’ shouldn’t dent your productivity too much.

One blocker to all this, of course, is other people. For example, it’s very difficult to move away from using Skype (which doesn’t have a web client) because it’s the de facto standard for business VoIP communication. That is only likely to change when there’s a critical mass of people familiar enough with different technologies to be able to switch to them quickly and easily. Hopefully WebRTC will expedite this process!

So, in conclusion, if you’ve got a workflow that depends upon a particular native app, perhaps it’s time to look for an alternative?* Then, at the minimum you’ve got that alternative up your sleeve in a pinch, and at best you may find you want to switch to it full time.**

 Image CC BY Robert S. Donovan

*For example, I’ve recently moved from Evernote to Simplenote and from Adium to IRCcloud.

**If you want to simultaneously focus on privacy/security, look at the newly-revamped PRISM Break site.

Want a tablet? Choose your vendor lock-in.

Ever wondered why Mozilla’s Firefox web browser exists? It’s because about 10 years ago Microsoft had sewn-up about 90% of the market and was creating vendor lock-in through anti-competitive practices. You can read about this in the History of the Mozilla Project. Happily, Mozilla were successful and now there’s at least two high-quality alternatives to Microsoft Internet Explorer – which itself has become more aligned with web standards. It’s a win for everyone who uses the web.

The next battleground is mobile. Although Google’s Android mobile Operating System (OS) is billed as ‘open’, for example, it’s not really developed in the usual Open Source way: the source code tends to be released long after each iteration of the OS. Apple, meanwhile, maintains a notoriously closed ecosystem with a stringent procedure for inclusion in their App Store. They also control how you can get things on and off iOS devices in order to make money from the iTunes store.

Amazon, meanwhile, is a fairly new to the mobile device game. They’ve taken Android and significantly modified it – including defaulting to their own app store. They’ve slashed the price of the Kindle Fire 2 (with, cleverly, ‘special offers and sponsored screensavers’) for Black Friday* making it a loss-leader. They’re betting on making the money back through Kindle book purchases, Amazon Prime subscriptions, and Lovefilm streaming.

So even though we may have multiple vendors it’s essentially similar problem to the Internet Explorer issue ten years ago. You may get shiny new ways to consume things that the vendor is selling you, but it’s not a great situation, overall.

You want a tablet? For Christmas 2012 that means you’re going to need to choose your vendor lock-in.

Thankfully, all this is set to change in 2013. Why? Two reasons. First, Mozilla are working on Firefox OS built entirely of standards-based web technologies. Secondly, Ubuntu Linux is being developed for mobile devices like the Nexus 7 and (even more excitingly) you’ll soon be able to run an entire desktop OS from your docked smartphone.

My conclusion? Buy a tablet if you have to, but be aware that real choice is around the corner…

(this was an attempt to write my version of the NSFW (but excellent) post by Terence Eden)

*The cynical nature of this marketing ploy is bad enough when tied to American Thanksgiving. It’s even worse when standing alone in the UK context.

Image CC BY-SA tribehut

On (not) working in academia.

On (not) working in academia

I’m with Doug Pete in really liking the Zite app.

Although it’s a proprietary, closed product I haven’t come across anything close to it for discovery. Take, for instance, a post entitled On Leaving Academia by Terran Lane, someone I’ve not come across before. He’s an associate professor in the Department of Computer Science at the University of New Mexico.

Terran is off to join Google.

His post is neatly organised into section titles listing the reasons he’s leaving academia to join Google:

  1. Opportunity to make a difference
  2. Workload and family/life balance
  3. Centralization of authority and decrease of autonomy
  4. Funding climate
  5. Hyper-specialization, insularity, and narrowness of vision
  6. Poor incentives
  7. Mass production of education
  8. Salaries
  9. Anti-intellectualism, anti-education, and attacks on science and academia

I’ve written about this kind of thing before in You need us more than we need you. As Terran explains, it’s not (just) about money.

Whereas he’s decided to quit academia, I’ve made a conscious choice from the start to stay on its sidelines. Around the margins. On the edges. Whilst the logical thing to do after my doctorate would have been to apply for a research position or lectureship at a university, I decided against it.


Not only would I be earning half the amount of money I am now – and less than when I was teaching in schools – but it seems a spectacularly bad time to decide to become a career academic. No money, no status, no freedom. And with the introduction of a market into UK Higher Education it’s increasingly difficult for academics to even claim the high moral ground.

That’s not to say academics aren’t doing good, publicly-useful work. Of course they are. It’s just crunch time in their industry.

I think we’re going to see a lot more of this talent-drain from academia. In fact, we’re already seeing a new generation of people not satisfied with traditional career structures and ways of working. I’m not sure if this is good or bad in the scheme of things, given the direction the universities (in the UK) seem to be headed under the current government.

What I do know is that universities need to do something, and fast. The Bank of Goodwill doesn’t have infinite reserves…

Image CC BY-NC-SA Patrick Gage

[UPDATED] Google+ Hangout about #openbadges TODAY 11:00 (BST)

Update: scroll down to video at bottom of post!

Open Badges - Google+ hangout

I’ve organised an impromptu, informal Google+ hangout for today about Open Badges (Friday 27th July 2012) at 11:00 BST. It’s in response to a few people on Twitter who wanted to ask me questions.

Most of those asking the questions were teachers.

If you’re interested, and you can make it, then head to the following URL just before 11:00 and we’ll talk through Open Badges until people run out of questions.


I’m using the ‘Hangouts on Air’ option so it should automatically be recorded and then available on YouTube afterwards!

There was an issue (explained in the video) so it’s up on Vimeo.

View Google+ hangout below!

What we talk about when we talk about Open Badges from Doug Belshaw on Vimeo.

[RECORDING] Connected Learning webinar on Open Badges

I was delighted to be asked to participate in a DML Central Connected Learning Google+ hangout about Open Badges yesterday. The recording should be embedded above, but if not try clicking here.

The session featured a presentation by Erin Knight, Senior Director of Learning at the Mozilla Foundation, and was facilitated by Howard Rheingold.

If you like this, you’ll also be interested in the webinar Erin and her colleague Michelle Levesque ran for the JISC Developing Digital Literacies programme last Friday. In that session, they discussed Mozilla’s work around web literacies.

Check that webinar out here, along with Erin’s write-up.

3 principles for a more Open approach.

This exchange on Google+ with Rob Poulter (referencing my previous post on platforms and standards) got me thinking. The highlights are below.


Ultimately I don’t think the problem is between native vs web, the problem is one of closed vs open, and not in a Google PR way. The things we tend to care about in the online world are services, not apps. Services see us passing responsibility for our data on to a third party, and usually based on features rather than interoperability or longevity. At the end of the day, if there’s something which we would mind losing, it’s our responsibility to keep it, not some third party.


My issue, I suppose is platforms becoming de facto standards because ‘everyone uses them’. Kind of like Dropbox and Twitter and so on…

There’s definitely an elision which I need to resolve in my thinking between ‘HTML5 webapps’ and ‘openness’. Thanks for the pointers!


The standards thing is tough I guess. Who wants to be the business that boasts of how easy it is to jump ship? Especially for social applications like Twitter, Facebook, G+ etc (Dropbox and other personal services not so much since they tend to compete on features and can’t rely on “hey, all your friends are here, you’re not going anywhere”).

I pointed out that Google Takeout actually does allow you to export your data from Google to other platforms. But, as Rob responded, not the comments on other people’s posts.

All of this made me think about my principles for using software and web services. It reminded me of Baltasar Gracian’s constant reminders in The Art of Worldly Wisdom (which I read on constant repeat) that it’s easy to begin well, but it’s the ending well that counts.

So, I’ve come up three principles to guide me:

  1. I will use free and Open Source software wherever possible. (I’m after the sustainable part of OSS, not the ‘free’ part)
  2. If this is not possible then I will look for services which have a paid-for ‘full-fat’ offering.
  3. I will only use proprietary services and platforms without a paid-for option if not doing so would have a significant effect on my ability to connect with other people.

What’s in and what’s out? I’ll stick with Twitter and Google+ (but will try to connect with people I follow in additional ways). Evernote, Spotify, Skype and Dropbox are fine for the time being (I pay for them). I’ll try and move away from GMail and Google Calendar.

Any suggestions for replacements?