I have gone through a series of iterations of my website and I tend to see them as grouped together like versions of a software release:
- 2000-2008 (versions 1.0, 1.1): These were HTML versions, mostly collections of bookmarks, based on a structure I designed and my friend Liam turned into a website for me. The site was mostly just pages of links for me and there was very little content on there for other people.
- 2008-2014 (versions 2.0, 2.1, 2.2, 2.3): These were variations to make the site work with Drupal as my content management system and I started expanding my content for others. Book reviews, movie reviews, and the start of posting my HR guide online.
- 2014-2017 (versions 3.0, 3.1, 3.2, 3.3): I switched over to WordPress for the blog and Piwigo for my gallery, and played with lots of options in between. I had serious issues with my hosting platform (GreenGeeks) and I finally gave up on them and moved to a new hosting provider (WHC.ca). Primarily, I worked on expanding content.
- 2017-2019 (version 4.0): I combined my two separate websites (ThePolyBlog.ca and PolyWogg.ca) into one and started expanding my content dramatically. I passed the 1000 posts and 1M words mark, and my site traffic jumped up to about 150 visitors a day.
Before the pandemic hit, I had been slowly chopping down the tree that was my old website. I was working on integrating my photo gallery, I changed some layouts and formats for things like book reviews, and there were a few issues I needed to fix here and there. When there were only a couple of hundred posts, little niggling problems weren’t much of an issue. But now that I’m over 1300, with some 28 categories and over 1.4 million words, those “little issues” are a bit more challenging. And this is, relatively speaking, the last chance I have to fix them before those problems become insurmountable if I need to go back and edit something. You’ll see as I go, none of them were terminal problems, but they did leave me with some major inconsistencies in my website. Some affected the front-end, most affected the back-end.
In the end, I decided to do all of them the “right” way, the way *I* wanted them to be, and the result is another jump in versions. I now have version 5.0 of my PolyWogg.ca site. Most of the work I’m mentioning below was stuff I did in the first six months of this year, I’ve just never summarized it before.
#01. Switching over to “blocks”
Two years ago, WordPress upgraded their back-end interface and released Project Gutenberg which replaced the classic editor with a “block editor”. In the past, you would basically type away and then if you applied formatting, you would usually do that afterwards. The “block editor” tried to move everyone one step closer to a desktop publishing layout tool, and well, let’s just say there’s a reason why the fifth most-used plugin across millions of sites is a tool that lets you revert your editor to the old one.
The short version is that when you go to start a new “section” on a page, you tell it in advance if you want it to be a paragraph, image, table, etc. The default is a paragraph, but whereas you used to just insert tables or images, now you insert a “table block” or “image block” and then add the table or image inside that “container”. For those used to doing desktop publishing, it’s the equivalent of drawing a box, telling the computer it will include an image, and then adding the image. Writers want a writing editor, which is what they had; now everyone from the writer to the final admin has a desktop publishing editor. No offense to the lovely folks who designed all this, there’s a reason why everyone uses Microsoft Word and not their companion app Microsoft Publisher to write their documents. But I digress.
Does anyone see a difference on the front-end? Nominally, no. At least not at first. A WordPress site would still show the code from either editor just fine. However, with the new blocks came some new power for users, and so designs started getting more complicated, more robust. For good or bad, the front ends of many sites started to change as people started to use their new tools. But viewers don’t “see” blocks, they just see the layouts changing.
For me and my site? Not so much. I hated the block editor. It was not intuitive at all to me, and many of the work processes I was used to using just didn’t work right in the block editor. So, like literally MILLIONS of other users, I installed the classic editor and just bopped along the old way. But I knew that I should eventually convert to the new method and I could see a few advantages, such as ways to replace a couple of inefficient work processes. Earlier this year, I sat in on the WordCamp for Dallas and when it was over, I fully committed to the Block Editor. I kept hearing the Mandalorian catch-phrases in my head. “They have spoken” and “this is the way.”
On the positive side, those 1300+ posts mean that I have a really good idea of what I need — and what I don’t. But I was committed to switching to Blocks, and I wanted to see what was out there…the good, the bad, the ugly, and well, the downright stupid ideas people had. The default Block Editor comes with a wide range of default blocks for everyone to use, but other companies have released dozens of collections of “extra” blocks with different “look and feels”. Many came with a lot of extra options to tweak them that the default ones didn’t — colour backgrounds, typography, ability to add images or icons, etc. In addition to the default blocks, I reviewed 9 other collections, trying out all of their blocks, and considered each and every one of the 209 possibilities.
No, you didn’t read that wrong, I looked at ALL of them. On the positive side, if it can be called that, many of the collections include similar options. For example, many included a special block for doing lists with the bullets replaced by “icons”. So 6 or 7 blocks that all do the same thing a little differently, and if you consider the overlaps, there are probably about 80-85 separate “types” of blocks.
One collection stood out way above the rest. Stackable’s collection generally did everything all the others did, and added way more tweaking options to every aspect of the block — text, size, spacing, colours, background, borders. Everything. They grouped it in three tabs — layouts, styles, and advanced. Most collections had small subsets of the styles and nowhere near the power of the advanced tab.
For the other collections, I tended to pick one or two out of their collection and disabled the rest. And there were lots that I just have no use for in my site. For example, I never show collections of old posts within another post or page, and there were quite a few designed to do just that step. So I simply disabled them.
In the end, I liked Stackable’s options so much, I upgraded to their Premium addition to get some extra sample styles and their support for a year.
I have disabled the classic editor, and I have gone all-in on using the block editor.
#02. Upgrading my featured images
Every time I post a new post, I assign it to a category and choose a “feature image” to go with it. For me, they are thematic. So all my book reviews should have the same featured image off to the side of the post. For those 28 categories I mentioned earlier, there are at least 28 images. But if I was honest with myself, sometimes I wasn’t really consistent. Suppose for example I write a book review about an astronomy book…do I use the astronomy image or the book review image? Is there a hierarchy in my mind?
Or what if I had goals related to astronomy? Should I use the goals images or the astro images?
While that is not a HUGE issue, I did have another challenge. I had added categories over time, and while not all the posts in that category used the same image, the bigger problem was that not all featured images were the same size. Some were 200px wide; some were 150px; some were 300px. I can override that in the site and “force” them to all be the same size, but they also weren’t all the same file format. Some were PNG, some were JPG. Some were transparent, some were not. Every once in awhile I would tweak something, it would look good, and then 3 weeks later I’d be looking at something else on the site and realize that my previous change had indeed worked for the image I had been working with for THAT day, but not for all the FIs. And suddenly I had things all messed up on a post that was a year old. I would “discover” it after the fact and think, “WTF? What happened to my site?”. I’d disappear down a rabbit hole only to emerge later with the answer that it was an old post that didn’t work well with a new tweak in all instances. It worked well enough, but messed something else up.
So I went hard-core. I reassessed ALL of my featured images, modified all my possible headers too, changed a bunch of images I use for signature blocks, and converted them ALL to transparent JPGs that are all the same size. No matter which image I choose, it is the same width as the others. I even had to accept that since dimensions/proportions change for some images that are taller than others, I went with a set width that limits the height to a set range.
But here’s the hard-core part. It meant replacing each and every featured image on every post. There are tools that will kind of do that for you, but not all of my posts had the same image to replace. And not all of them by category would have 100% the same image. There is SOME variation. Which meant the best way to do it was to edit them all individually.
To replace images in a way only I would probably notice?
That sound you hear is crickets. Even *I* didn’t think that was a good enough reason.
#03. Fixing inconsistencies in some common elements
Consistency is the hobgoblin of little minds, and editing old posts on a website might be the work of the littlest mind of all. But I didn’t like that a couple of things I had done over time didn’t really look right.
At the end of every post, I have a signature block. My default one says “Happy reading, PolyWogg”. I really like the tool, I created the block in PowerPoint, and I love the tagline to close my posts. I don’t know why, I just do. But then I added some for TV show and movie reviews, and “happy reading” didn’t seem quite right. For astro posts, the more common salutation is “Clear skies”, and I added that one. But I have 28 different categories, and when I was done reviewing them, I figured I might need up to 10 separate signoffs.
Sure, I could do that starting now. Just upload, use the new ones from here on out. Easy peasy lemon squeezy. After all, how many of the old ones really needed editing? 10%? Not worth it.
#04. Changing my post layout
I use a theme that I have used for over 10 years in one form or another. It works AWESOME for me. Weaver Extreme. There are more powerful options now, but this one just gives me tweaking options out the wazoo and I still love how it looks.
Except for one thing. It wasn’t big, just a “I wish…” moment. At one point, I was going through a bunch of sites, and I saw a bunch that had their dates sitting outside the post i.e., to the left of the text, and it looked pretty sharp to me. There is an option in the theme to include your featured image out there, but I always embedded it in the page and wrapped text around it. But, being able to set out the date? I couldn’t do it with my theme and so I pushed it aside, no sense crying over missing milk.
Or…could it do it? I went to the support forums for Weaver, and connected with Scrambler who is a Weaver user extraordinaire and offers support in the forums. She or he has helped me in the past with other glitches or tweaks, and I was happy to see they were willing to help with what I wanted. I confess, I didn’t even really know where to start. I wasn’t sure if it was a theme setting, HTML code or simply CSS code. I didn’t even know if it was possible, but I described what I wanted and they said it was possible with CSS code. Hot damn.
If you’re reading this on a desktop, go back up by the title and you’ll see that the featured image is set off from the main prose and below it is a date styled with red and white boxes that looks a lot like the classic Google calendar boxes. Scrambler helped (i.e., spoon-fed) me in figuring out the right code to make it happen. Then they told me how to fix it so it didn’t break bad on mobile and tablet.
But with that change made, and the FI sitting out there too, that problem I mentioned with all the FIs not being the same size and proportions started being something larger to consider changing.
Back in the day, mobile-friendly sites required you to have two separate themes running concurrently. One handled desktop, one handled mobile. If you loaded the site “m.yourdomain.com”, it ran the mobile theme; if you loaded, “www.yourdomain.com”, it loaded the full site. My theme was mobile-friendly in that sense, but over time, it kind of stopped as the mobile tools upgraded and I didn’t. The new approach is to make your pages “responsive” — they shift in size depending on the size of the screen viewing it. Any theme WILL change, but responsive means it changes intentionally and collapses gracefully.
Someone had mentioned in a Reddit thread that they were trying to access a number of pages in my HR guide (you know, the main draw to my site), and the pages weren’t very mobile-friendly. Why? Because they used a lot of tables. Damn it. I liked tables. That’s why I used them.
Or was it the reason? Hmm…if I thought about it a bit more (and I did), I realized I used tables because it was the simplest and fastest way in the old editor to control the alignment of elements. But the block editor has a LOT of different blocks that help with multiple types of alignments.
I could start using those today, but if I wanted to improve my existing friendliness, I would need to edit a number of posts and pages.
Nope, I was still going to resist going back to old posts and updating them.
#06. When is a meme not a meme?
I have my HR guide, that’s a big draw. People love my posts on my telescope alignment tips. A few other posts here and there get a lot of hits (Phoenix audits anyone?). But none of those are really why I started my site way back when.
I started it to post my book reviews, movie reviews, humour and quotes.
I have often played with layouts of all of them, but the two biggest challenges to “gel” as a concrete layout are the humour and quotes. I love Aunt Acid layouts where they have a common image, a few lines. E-cards are another layout I like. So I’ve played with lots of techniques to design simple “cards” or “memes” where I have some text, common branding with my frog logo, and a joke or a quote. Perfectly designed so someone can “share it” as a graphic. Like a meme machine.
Yet here’s the rub. If I post a great quote in a graphic, the website doesn’t “see” the text. It doesn’t index the words, or the source. Sure, it looks pretty, great for sharing, but if someone searched for a quote by Abraham Lincoln, my site would NOT pop up anywhere. For all intent and purposes, my site is invisible to search engines for that content. By contrast, if I just put the text in the site, then it isn’t a pretty graphic that can be quickly shared. Some people put in the pretty image and add the text as ALT-TEXT. That is a LOT of extra work in my view.
Finally, after hemming and hawing between text or graphics, I bit the bullet. If I have to choose between having the text on my site and having a pretty picture to share, I’ll take the text. I want to be able to find it easily myself. So I don’t have the meme option? At least I have the text. And yet, I can put a pretty picture behind it on the website, it’s just not shareable as easily. Big deal.
Oh, except to convert my existing ones over meant going in and editing all those posts individually.
The universe was very clearly telling me to fix / edit with more consistency and so I took everything off-line and started editing the 1300+ posts, fixing tons of stuff as I went. Layouts, FIs, signature blocks, tables, etc. And in particular, all the special layouts I have for book reviews, movie reviews, TV reviews, humour and quotes.
Let’s see what I have for posts, noting of course that some occupy more than one category so it isn’t simple addition:
- Astronomy: 94 posts;
- Book reviews: 252
- Civil service: 55
- Computers: 127
- Development: 28
- Experiences: 176
- Family: 115
- Goals: 254
- Governance: 31
- HR Guide: 13 (although most were moved to pages)
- Humour: 86
- Ideas: 64
- Learning: 111
- Libraries: 8
- Movie reviews: 12
- Music reviews: 17
- Photography: 54
- PS Transitions FP: 1
- Publishing: 30
- Quotes: 96
- Recipes: 19
- Spiritualism: 17
- Television: 267
- Writing: 51
I’d love to say it all went swimmingly, but it didn’t. Some of the pieces were significant challenges. One of the blocks I was using was causing wonky issues on my phone — it kept thinking my background image was actually a video so it was trying to give me a link to “play it”, but since it wasn’t actually a video, it was giving me the play button with a big line through it saying it couldn’t play. Well, no sh**, Sherlock, cuz it’s not a video. Stackable realized it was an actual bug when I submitted my support ticket, AFTER I spent a ton of time with Scrambler teaching me how to handle it a different way. Then Stackable fixed the bug, and all the glitches went away and I went back to using it normally. Nice.
I basically went category by category, as most of the “issues” were common across themes:
- Adjusting some nomenclature in the titles (simple things like where I used dashes in one title and colons in another, or a series of articles that I reviewed and some said “Articles I Like:” as a category, and some didn’t);
- Switching from classic paragraph block (one big block) to separate blocks to allow better editing…if I didn’t have to do sub-edits, I just left the classic paragraph one running;
- Replacing the signature block with the new ones;
- Replacing the featured images with the new transparent ones that are all the same size;
- Tweaked the size of the social media icons after each post;
- Leaving the “big” format posts to the end – recipes, book reviews, music reviews, movie reviews, humour, quotes, and TV reviews.
I also was able to create a bunch of reusable blocks to ensure consistency in layout and format in the future, so that I won’t have to redo any of this again. And, to be brutally honest with myself, in a way that if I do decide to do something different in the long-run, I can leave the current one as is and just start “fresh” from the new point. I don’t need to be in lockstep for the past and future posts, just improved flow and consistency for the above elements.
Will anyone else notice? Probably not. Some people have noticed a few improvements here and there, but most of them won’t see a dramatic shift — it’s still the same theme and colours.
#07. Fix the admin menu on the back-end
Nobody else sees the admin back-end menus but me, but I have a bunch of plugins and the order has always been a bit chaotic. Things that I thought were low priority were near the top, others that I use almost daily were buried at the bottom. That’s just a feature of WP, it lists them in the order they were installed or as per whatever order the developer thought was good. Fortunately, I could use an advanced admin menu editor and go in and rearrange everything the way *I* want it — posts / comments / pages/ gallery and media / downloads all near the top. Plugins / settings / appearance / blocks all near the middle. And users / tools / security near the bottom. Pretty much in the order of the frequency with which I use any one section.
#08. Auto sharing of posts on social media
When I write a post, I usually finish it, save it, preview the final version on the website, and then I manually share it separately on social media (well, FB and Twitter). The “best practice” in the industry is to auto post to social media so you don’t have to repeat those steps, and up until 3 years ago, that was easy to do. WP and FB worked fine together. Then FB closed some posting privileges for individuals, and suddenly you couldn’t auto-post to FB. I changed my approach, did most of it manually, and relatively “sat on it”. It wasn’t a big problem for me, but it was a small niggling issue. Particularly when I write book reviews where I end having to share it 3 different places!
I downloaded SNAP’s auto poster and sure enough, FB didn’t want to play ball. They advertised that if you went with the paid version, you could enter your login and password and it would work perfectly. Yeah, no. I paid and what you ACTUALLY have to do is try about 30 times to get it to accept the FB sharing with your account, and each time, it kicks you out of FB and makes you log in again. It’s supposed to do that 4 or 5 times and then “accept” it after that, but I was well over 30 at one point and no dice. Lots of reviews online say the same thing in much harsher terms for their false advertising. However, you CAN do a back end tweak to fake your way into letting it share, and if you just leave the cookies and sessions running (even if you close the browser), it’s all good. As long as you don’t “log out”, it mostly works.
I use it sporadically, but it saves me a few steps, sure.
#09. Re-organize my HR Guide posts and pages
I mentioned above that my HR Guide is the most popular page on my site, and there are sub-Reddit threads devoted to discussing it. Where users have posted links to pages they liked. And therein lies the problem. A few of those pages are ones that I’ve “archived” as old content, but I left them “live” while flagging very clearly that they are OLD with links to the NEW versions. Yet because people see the links on a Reddit link, they end up jumping there when they click, and they frequently don’t follow the link to the latest version. AND they often often pose questions in the comments on the old pages. WTF?
When I turned off comments on the archived pages, people complained they couldn’t leave comments on a dead page. WTF x 2?
In the end, I put ALL of my content on a page whose URL won’t change, which was the obvious solution, but it left me two issues.
First, I didn’t want people just getting error messages on the old pages. So I went through and set up formal redirects to send them to the RIGHT page with my table of contents for the latest elements.
However, the second issue wasn’t as easy…people had detailed comments and replies on those old pages. Content I didn’t want to LOSE when I archived the page. Fortunately, I was able to find a plugin that lets you move comments from one post to another. I was a little skeptical at first, but the plugin worked like a charm. Everything moved where it was supposed to go, no hiccups.
What I really have been wanting to do over the last few months is to re-write the guide and update it to 2020, but the revamp of the site had to take precedence, so I haven’t got past Chapter 3 yet. It’s still on my list, but I’m a bit disappointed that the site revamp took so long to do it “right”. I was hoping to have the HR Guide all done by my birthday which was back in June. It wasn’t critical, an artificial deadline so to speak, but it would have been nice.
#10. Limit tables where possible to TablePress
I mentioned earlier that tables are generally not “mobile-friendly” and the best practice in design is not to use them at all. But there are times where a table is just the simplest and best way to present data.
Take my various reviews for instance. I have about 190 book reviews. And to help people find them, I used to have a bunch of manual lists that I created so that it WOULDN’T be in a table. They were separate lists and sorted by:
- Book title;
- Author last name;
- Review number;
- Date of the review;
- Year of publication;
- Series name (if applicable);
- Rating (out of five); and,
- The link to the review, of course.
It’s a lot of information, I keep it and more in a MS Excel spreadsheet, and I was adding it to manual lists on seven separate pages. It was a pain in the patootie.
Enter the plugin TablePress. It lets you store all the info as if it was a spreadsheet like file, and you can even “paste” from an Excel spreadsheet into the back-end to generate the data. No need to retype it. But that’s not the big time saver.
Once you generate the table, and it does have to be “generated” to see it live, you can set it so that ANY column is completely sortable on the page. One page, 7 columns, all sortable. No waiting. No repasting. No editing. If you want to sort on the title, click the column. If you want to sort on the author, click the column. Easy peasy lemon squeezy.
And with a premium plugin, I can even make it “filterable” in advance plus add a search box to the top of the table that lets you self-filter the table. Only want the entries that were published in 1998? I can prepopulate it that way and JUST show you 1998, or you can type it in the search box yourself. It’s awesome.
It turns your simple table into more of a searchable database. It’s not perfect, but it is WAY more power than I had. With the only “hitch” being that it is a table. It doesn’t look great on a mobile phone. But you can search on the main page just fine with a phone, so if they can’t work the table for my book review index? No huge loss.
I have a few other pages that will also have tables on them, but there’s no getting around listing data that way. I could do a LOT of extra work for minimal benefit, and it wouldn’t look right any other way. Instead, I’ll sacrifice a bit of mobile-friendliness to save myself a ton of work.
I did a bunch of other stuff too, little tweaks here and there…I reduced the number of active plugins and cleaned up some outstanding “security” tweaks I wasn’t sure if I should make permanent or not (I did).
I still have about 50 posts to fix before I am back fully live, but that is more a question of a huge design question for a small problem with small single images from my NextGen Galleries. I solved it, I think, but that is worthy of its own post all by itself. And I need to tweak my approach to both backups, photo management, and a simple glitch with my CRON (automatic macros) scheduling. Nothing challenging, but they need to be done.
Well, I shouldn’t pooh-pooh the photo management. I have photos embedded in posts from my old photo gallery, and once I get them all embedded properly in my main site, I need to “update” those links. So, as I went, whenever I saw something that would have to be “updated” later, I added a flag to the post. It is live, but I can pull it up with a simple sort. That’s part of a much larger photo management project, but it means between 145 and 185 posts need to be updated at some point. Once I get to that point, it’s a few hours work, but it’s still outstanding.
With the amount of work I did above earlier in the year? I can live with those items still being on my to-do list.