Some recent things I have learned about SEO

A screenshot from Google Console Search Insights congratulating me on achieving 900 clicks in 28 days, indicating that I am doing okay from SEO perspective

I’m signed up to Google Search Console (formerly known as Webmaster Tools) which gives web site owners an indication at how good they are at SEO, or search engine optimisation. And last weekend, Google emailed me to congratulate me on 900 click-throughs in 28 days. That means that about 60 people find something from this blog in a Google search, and click through to see it.

Of course, I mainly blog for myself. It’s a chronicle of things that I have done, or things I’ve found useful or want to share. But I also want people to be able to read it; as the saying goes, if a tree falls in the forest and no one is around to hear it, does it make a sound? So I feel it’s important to incorporate some degree of SEO.

As this is a WordPress blog, I use the Yoast SEO plugin (the free version). It works okay, and gives each post a traffic light symbol for SEO and readability. Readability is important to me; it discourages me from using overly long sentences, and avoiding passive voice. But the SEO suggestions are helpful too. For example, making sure that any images have relevant alt text, and that I include both internal and external links. Not every blog post needs to be optimised, but I try to optimise most of them.

Some blog posts have done better than others, and seeing as the whole idea behind this post is that I share what I’ve learned, here are some examples:

Comparing things

People will often search for ‘[thing 1] vs [thing 2]‘, which I found out when my blog post about the films Wonka and Wish did well back in January. It just so happened that I had seen both, and I could have reviewed them in separate posts. But putting them together meant that my little blog appeared on the first page of results for a while. Alas, apart from watching Butterfly Tale last month, I haven’t been to the cinema since.

I put this into practice when comparing Readly and Pressreader. My original plan was just to review Pressreader on its own, but having seen how well my Wonka and Wish post did, I tried comparing the two. That blog post is typically in the top 5 results for the search term ‘readly vs pressreader‘.

How-to guides

The most clicked-on link from Google is my guide to installing Home Assistant Supervised on a Raspberry Pi. About a third of all clicks to my blog are to that post alone. But other how-to guides score highly too, including Mounting a USB hard drive on startup on Ubuntu Core and How to join a preferred Thread network in Home Assistant. Generally, I’ve written these guides because I haven’t found a simple answer myself, and these are usually an amalgamation of advice taken from various forums or StackExhange threads.

Whilst having a dedicated SEO plugin probably helps, there are some things that you can do yourself that can help:

  • Ensure you have a machine readable site map, and point Google at it.
  • Make sure your site loads quickly; one of the reasons why I changed to GeneratePress as my theme is that it’s a lightweight theme.
  • Post regularly – Google likes sites which often post new content. I’ve found every other day to be manageable.

As I said before, the number one person that this blog is written for is me. But I also want it to be useful and for people to find it. I hope I get the balance right.

Back to posts every other day

When I re-started blogging regularly again, my aim was to post a new blog entry every other day – so 3 or 4 new entries per week. Over the past couple of weeks, I’ve been posting daily. This is because:

There was no new blog post yesterday, and there won’t be tomorrow, as I’m now going back to posting every other day. Whilst I still have lots of blog post ideas, if I try to post a new blog entry every day, I’ll burn through them quite quickly. And now that I’m back at work full-time, I won’t have as much time to write. I’ve tried having a new blog entry to post every day in the past, and it’s hard work.

22nd blogiversary

An AI generated image of a birthday cake with two candles on it showing 22

It’s been 22 years to the day since I wrote my first blog post. Which is a very long time in technology.

Last year, I referred to the blogiversary as a ’21st-ish blogiversary’, mainly because I took a break from blogging for four years and all my pre-2018 where lost. However, I’ve been able to reinstate the first one, so you can read what 17 year old me had to say about starting a blog.

For the first few months of blogging, my posts were made using Blogger, which I’m almost surprised is still around. Google bought Blogger over 20 years ago, and hasn’t done much with it. It’s still there, and you can still create a new blog, but apart from a new responsive UI in 2020 it seems to have stagnated.

I switched to Movable Type in September 2002, and bought the domain that I’m still using today. Over time, I switched web hosts a couple of times, but I’ve been with Bytemark for many years now.

In March 2011, I moved from Movable Type, which was becoming a commercial product, to Melody, a community fork. Alas, there just wasn’t much interest elsewhere in the blogosphere in either Movable Type or Melody, and so this was a short-term change. Just a couple of months later, I switched to WordPress, and have been using it for almost 13 years now. That’s longer than any of the other blogging systems that I have used in the past.

Over this past year, I have been trying to get back into blogging, and, since October, I have (for the most part) published a new blog post every other day. I’m enjoying blogging again and wish I’d re-started sooner, although at least with a 4 year break, there’s plenty to talk about. I still have plenty of ideas for future blog posts which should see me well into the spring. So, happy birthday blog – you’re looking better than you have done in years.

WordPress backups with UpdraftPlus

A meme featuring Anakin and Padme from the Star Wars films. Anakin is saying 'I've reinstated my WordPress install after everything got wiped' and Padme says 'so you've got backups now right?'

It’s a little while until World Backup Day on the 31st March, but I’ve set up UpdraftPlus to create automatic backups of my WordPress installation to my Dropbox account.

It’s a straightforward plugin to set up. You install it from the WordPress plugin directory, select your cloud storage provider, choose what you want to backup, and then run it. If it all goes well, you can then set a schedule for automated backups.

The free version offers Dropbox, Google Drive, Amazon S3 and some others, and you can also upload your backups to any FTP server, albeit over a non-secure connection. If you want to use WebDAV, SFTP, SCP, Microsoft OneDrive or UpdraftPlus’ own service, then you’ll need a premium account. At present, it’s £54 per year for up to two personal sites, which is pretty reasonable – the equivalent of a little over £4 per month.

Your backups can include everything if you want, but I’ve excluded plugins and themes from my backups. I haven’t modified the theme that I am using, and I’m not using any custom plugins – everything is from the WordPress plugin directory.

Backing up my files to Dropbox makes the most sense to me. I pay for Dropbox Pro and so have 2 terabytes of storage, of which I’m using less than 7%. Whilst VaultPress is the officially-supported solution for backups for WordPress, it’s not a free service and I’m already paying for Dropbox Pro.

The best time to set up backups is now

There isn’t a ‘best time’ to set up backups for any system apart from, well, now. So, if you haven’t got a backup solution in place, this is your reminder to sort something out. I lost all my blog posts in 2018 because I didn’t have adequate backups in place when doing a server upgrade. I would say I learned my lesson, but I’ve been blogging again for 18 months now and only set up UpdraftPlus last week.

There are many other WordPress plugins that offer backups – some free and some paid. I chose UpdraftPlus as it seemed to be the one which offered the features that I wanted, but you may find another suits you better. Just make sure that you have something in place.

New old posts from the archives, part 3

I’m gradually bringing back some of my old blog posts that were lost. You can have a look at part one and part two, and here’s the list of latest posts that I have brought back:

  • WeeMee (September 2003). I mentioned this in last week’s blog post about avatars, hence why I brought it back. The site that lets you generate a ‘WeeMee’ is long gone, and I’m not sure that an archived copy would work as, if I remember correctly, it used Adobe Flash Player.
  • Secret Starbucks Sizes (July 2015). This seemed to show up quite a bit in my 404 (missing page) logs and so I brought it back. It lists the sizes of coffee cups that Starbucks may or may not offer besides their usual Tall, Grande and Venti sizes.
  • Coffee (February 2012). Linked in the above blog post, so it made sense to bring this one back too. I never used to be a coffee drinker until around 2012. Now, I drink coffee most weekdays, although it’s usually a sachet of pre-mixed powdered coffee with a flavouring. Starbucks is a once a week treat on the way to work.
  • My Podcast Diet (August 2018). Well now, this is a post that I never finished; it was sat as a draft on my iPad and didn’t get published. So you can now read it for the first time. I should probably write an update soon, as I’m listening to some different podcasts now.
  • Going into print (August 2004). A prime example of announcing something before it’s ready. I agreed to contribute a couple of chapters to a book about Movable Type, the blogging system that I used to use before switching to WordPress. And whilst I did start writing them, it clashed with my final year at university, so the book was published without my contributions. You can still buy the book (sponsored link) but no-one uses Movable Type any-more so I don’t know why you would want to. Almost 20 years later and I still haven’t co-authored an actual printed book.
  • Why I’m not switching to WordPress (August 2004). Oh wow. This was a 1000+ word rant that WordPress, barely a year old and only at version 1.2, lacked many niche features that I was accustomed to in Movable Type. It even garnered a response from Matt Mullenweg himself, which you can read if you’re willing to track it down on the Web Archive. Of course, less than a decade later, I would switch to WordPress, and Movable Type is basically dead as previously mentioned.

The Birthday Gift of Blogging

An AI-generated image of a blog post popping out of a gift box whilst a crowd of people look on in awe

You may have heard of Matt Mullenweg – he co-founded WordPress, and is now the CEO of Automattic, which owns and contributes to (the downloadable version that you install on your own server, like I do).

Matt, like me, is turning 40 this year. And, as a ‘birthday gift’, he has asked people to blog. About anything.

Having only recently returned to blogging regularly, I’m starting to enjoy it again. With the rise of the Fediverse, the web is starting to feel like it did again in blogging’s heyday in the mid-2000s. Whilst some people used centralised sites like Blogger and TypePad back then, you could use tools like WordPress and Movable Type to run your own blog on your own server, and still interact with everyone else. And then the closed gardens of Twitter, Facebook, Instagram and the like came along, and you had to be on those platforms to interact with other users.

Anil Dash has written a piece for Rolling Stone magazine called ‘The Internet is About to Get Weird Again’. And I think he’s right; there’s growing dissatisfaction with these big centralised services and how they hoover up so much personal data to sell to advertisers. Meta’s Threads is dipping its toes into the Fediverse, which would mean that Threads and Mastodon users can follow and interact with each other in a way that hasn’t been possible before. And if it works for Threads, could we see Instagram joining the Fediverse, to connect with PixelFed and maybe even Flickr?

Anil has some further reflections on his own blog. I’m hesitant to say that ‘blogging is back’; after all, there are so many other places that we can share short thoughts where there are audiences. But I feel like it’s having a bit of a renaissance, and in an age where there’s so much AI generated waffle filling up our search engine results, being able to interact with other humans has never been more important. It was what the web was designed for, after all.

So, happy birthday to Matt – and here is your gift from me. I hope that many others will do the same.

AI art

An example of AI art - a generated image of a robot painting a picture of some flowers on an easel

You may have noticed that several of my recent blog posts have featured custom AI art work related to the topic. I’ve generated these using Microsoft’s Bing AI Image Creator, which uses OpenAI’s DALL-E text-to-image model. DALL-E can generate an image based on a text prompt; for example, the featured image on this post was ‘A 1950s style robot standing in front of an easel painting a bouquet of flowers in a vase’.

These are some of the other AI art images that I’ve used recently:

AI art is controversial. It can create images in a few seconds that would take a human artist hours or days to produce. And, in some cases, these image prompts can be told to create images in the style of a particular artist, depriving them of income from a commission. It’s also notable that models like DALL-E and Stable Diffusion have been trained on copyright works, without the rights holders’ permission.

With this in mind, I’m justifying my use of AI art on some of my blog posts because I’m not an artist myself, and as an individual blogger who doesn’t make money from blogging, I wouldn’t have the money to pay a human artist. Whilst I have over 5000 photos that I’ve uploaded to Flickr, there isn’t always a relevant photo to use that I have taken. For example, in my recent post on comment spam, I decided to generate the above image of a robot converting blog posts into a tinned meat product, because I don’t have a photo that represents that. And whilst I make use of screenshots where relevant, sometimes this isn’t appropriate.

Of the AI art generators that I have used, the Bing AI Image Creator seems to be the one that gives me the best results. Any images you create are saved in the cloud, and can be downloaded for re-use. And each prompt produces four images so that you can choose the one which looks the best.

More new old posts from the archives

An AI generated image of a phoenix rising from the flames of a browser window

I’m gradually bringing back some of my old blog posts that were lost, and here are links to the latest batch that I’ve made live again:

  • Create a Safely Remove Hardware shortcut (April 2007). Another of my how-to blog posts, this allowed you to create a desktop icon that opens the Safely Remove Hardware function in Windows. Surprisingly, this still works in Windows 10.
  • Screenshots on a PocketPC (November 2005). How to take screenshots on a PocketPC/Windows Mobile device, since there wasn’t a built-in screenshot app.
  • Resurrecting a dead OS with KernelEx (May 2010). KernelEx is a compatibility layer for Windows 98 and Me that allowed apps which would normally require Windows XP to be installed. You can still download it, but it was last updated in 2011 and I expect there’s not much demand for it nowadays.
  • My favourite add-ons for Thunderbird (May 2014). Some add-ons that I used with Mozilla’s Thunderbird email client. I don’t use it anymore – we use Outlook at work and at home I tend to just use Gmail.
  • How to: get cheaper car insurance (September 2015). Another how-to guide. When reposting this, it was surprising how few sites back in 2015 used HTTPS by default; Let’s Encrypt had launched the previous year so I’m guessing Google hadn’t yet started favouring HTTPS sites in its search results.
  • A car. An actual car (September 2015). Linked with the above, the purchase of our first car. Sadly said car got written off in France in 2019, although it subsequently got back on the road with a new owner.
  • Passed (August 2015). Again, linked with the above, this was about me passing my practical driving test.
  • Expecting (July 2015). The announcement that my wife Christine was pregnant. Linked from the above posts, but I am also looking to reinstate blog posts about major life milestones as well.
  • 20 week scan (September 2015). As above.
  • Back in the driving seat… (July 2014). Restarting driving lessons, which led to me passing my test, as above.
  • Theoretically passed (April 2015). Passing my driving theory test. A lot of the above posts all link together, and so I’ve wanted to bring them all back at once to avoid creating dead links.

There are, of course, more to come. Whilst I estimate that I’ll only be bringing back around 10% of the old blog posts, that does mean around 250 posts to copy from the Web Archive and update.

Comment Spam strikes back

An illustration of a robot turning web pages into canned meat product. Generated using Bing AI Image Generator

So now that I’m blogging again, it’s the return of comment spam on my blog posts.

Comment spam has always been a problem with blogs – ever since blogs first allowed comments, spam has followed. Despite the advert of the rel=”nofollow” link attribute, automated bots still crawl web sites and submit comments with links in the hope that this will boost the rankings in search engines.

In the early days of blogging, blogs often appeared high in Google’s search engine results – by their very nature, they featured lots of links, were updated frequently, and the blogging tools of the time often produced simple HTML which was easily parsed by crawlers. So it was only natural that those wanting to manipulate search engine rankings would try to take advantage of this.

I’ve always used Akismet for spam protection, even before I switched to WordPress, and it does a pretty good job. Even then, I currently have all comments set to be manually approved by me, and last week a few got through Akismet that I had to manually junk.

Humans, or AI?

These five interested me because they were more than just the usual generic platitudes about this being a ‘great post’ and ‘taught me so much about this topic’. They were all questions about the topic of the blog post in question, with unique names. However, as they all came through together, and had the same link in them, it was clear that they were spam – advertising a university in Indonesia, as it happens.

Had it not been for the prominent spam link and the fact they all came in together, I may have not picked up on them being spam. Either they were actually written by a human, or someone is harnessing an AI to write comment spam posts now. If it’s the latter, then I wonder how much that’s costing. As many will know already, AI requires a huge amount of processing power and whilst some services are offering free and low cost tools, I can’t see this lasting much longer as the costs add up. But it could also just be someone being paid using services like Amazon Mechanical Turk, even though such tasks are almost certainly against their terms of service.

I think I’m a little frustrated that comment spam is still a problem even after a few years’ break from blogging. But then email spam is a problem that we still haven’t got a fix for, despite tools like SPF, DKIM and DMARC. I’m guessing people still do it because, in some small way, it does work?

Bringing back the archives

An illustration of a phoenix rising from the ashes, with a web page. Generated by the Bing AI Image Creator

Last month, I wrote about how I had found peace with myself regarding losing over a decade’s worth of blog posts.

Well, I’ve already sort-of changed my mind. I have already brought back some old posts which, until now, were only accessible on the Web Archive Wayback Machine.

This doesn’t mean that all of my old posts will be reinstated – if anything, I’ll be bringing back 1-2% of them at most. My criteria are:

  • Posts which, despite being offline for about 5 years, are still linked to. I’m using the Redirection WordPress plugin to track 404 errors, which you can group by URL to see the most commonly not-found pages.
  • Posts that still offer useful advice, or information that is otherwise not easily accessible on other web sites.
  • Posts that mark important events in my life.

So, here’s a selection of what I’ve brought back already, in chronological order:

  • Media Player Classic (January 2004). A review of a now-defunct lightweight alternative media player for Windows. VLC is probably a better option nowadays.
  • Apple Lossless Encoder (May 2004). A blog post about Apple’s then-new music format which preserves full audio quality when ripping CDs in iTunes, and how it compares to other formats like FLAC and Monkey’s Audio.
  • Knock-off Nigel (August 2008). An anti-piracy advert for TV.
  • How to migrate a Parallels virtual machine to VirtualBox (November 2008). A how-to guide for switching from Parallels Desktop to VirtualBox, which I imagine is still useful for some people.
  • Fixing high memory usage caused by mds (February 2013). A how-to guide for fixing an issue with MacOS. I don’t use a Mac anymore but hopefully this is still useful to someone.
  • Baby update (November 2015). This was actually a draft version of a post that must have somehow survived in Firefox’s local storage, so I re-published it.
  • How to: fix wrong location on iPhone (January 2017). Another how-to guide that fixed an issue I was having at the time with my iPhone’s location randomly jumping around.

There’s more to come, as and when I find time to restore them. I’m also using Google Search Console to find pages that it’s expecting to work, but that result in a 404 error.