This certification was interesting in a few different ways – first, it was kind of done “spur of the moment”; I’d passed my other exams and wanted to keep the momentum going, so I booked it without much forethought. Secondly, I’ve done loads of Secure Software Development Lifecycle training and documentation over the last few years, so I feel I might have had a leg-up on at least some of the fundamentals of this topic (pardon the pun). Finally, I genuinely found some of the tools referenced in the training to be quite interesting in and of themselves – something I hadn’t appreciated before diving in.
I’m planning to have a bit of a break before I’m back on the exam trail; I next have some virtual classroom training in September and October, followed by Associate-level exams in October and November, which will be my last for the year.
Sortly after this screenshot was taken, my office hit 31.5°C, which is about the time I decided it was time to stop for the day.
If this is going to be the norm for the next few weeks, I’m going to have to find some way of cooling down the room that’s better than my current option of “open the window”. Scots aren’t built for this sort of heat!
Now that I’d completed the data side of Azure, I wanted to try out something I was a little less familiar with; I’ve not gone near anything Machine Learning or AI-related in my work so far, so I felt getting at least some exposure to these topics wouldn’t go amiss. I’m glad I opted to add Microsoft Certified: Azure AI Fundamentals [AI-900] into the mix of what I’m studying this summer. It was a fun course, and one I fully recommend.
I wasn’t expecting to enjoy this topic as much as I did. Maybe it was partly a reaction to how little I’d enjoyed the Power Platform Fundamentals course. AI/ML is not an area I see myself working on often, possibly never outside of certification courses, but I don’t think I’d mind at all if someone did ask me to jump into a small project using Azure’s AI platform and tools.
A lot of the exam focused on the Responsible AI priciples, along with identifying the different types of workloads, tools, and models to use. This was one of the easier exams I’d sat lately. It took me around twenty minutes to complete, and I passed with a higher score than I expected.
As far as resources go, I used my regular mix of Microsoft Learn’s free study material, and Whizlabs. The prep for this exam was a little more practical than some of my other exams lately, which is probably why I found it a bit more fun.
It’s been a very long time since I visited the art gallery. Almost a decade, in fact. When the refurbishment was completed in late 2019 I heard a lot of good things, but before I could get around to paying a visit, COVID hit and we were all locked indoors.
With the easing of restrictions, we were looking for a quiet something to do on a day off – and as Caley had never been, the Art Gallery seemed an ideal choice.
The visit started off well; I walked into what used to be the entrance, only to find it’s currently the exit due to the one-way system they have in place as a safety measure. But once we got that snafu, things went well.
The gallery is so much more light and airy than I remember, with more natural light now. It feels a lot bigger, with more to see (even though half of it was closed off to prepare for a new exhibit). There’s a lot of subtle marrying of old and new throughout which I liked. I also really liked the new rooftop terraces, as there’s some nice views can be seen (and I’m sure more to come once Union Terrace is completed).
Overall, I was impressed, and I’m sure we’ll be scheduling more regular visits in the future.
To the side you’ll see a selection of the photos I took throughout our visit. I was consciously trying to avoid experiencing the visit just through the viewfinder, so there’s not a huge number of photos, but even though, this is only a small selection of what I could have shared.
Let me be upfront by saying: I really did not enjoy this certification. For whatever reason, I just could not connect with the subject matter, and the last few weeks have felt like an uphill struggle the whole way through. Even when I tried getting hands-on with the various pieces of the Power Platform, a lot of it was just plain unenjoyable. Power Automate and Power Virtual Agents seem interesting enough, but Power BI and Power Apps are areas I’d be happy to not have to be around again.
Maybe it was because I didn’t know much about Power Platform before starting (other than a nugget of Power BI exposure in my Azure Data Fundamentals certification), but I really did not expect so much of the course and exam to be taken up by Microsoft Dynamics. At times it felt like a big disadvantage to not have prior knowledge and experience of Dymanics and Dataverse.
The exam itself was fairly straightforward. It felt like the natural follow-on to the general Microsoft Azure Fundamentals (AZ-900) certification I passed last year, and follows the same multiple-choice/select-the-right-phrase format as AZ-900. For the most part it meant knowing which Azure data service should be used in a given scenario, or what feature in a solution fulfils the need; for the most part it’s usually straightforward to spot, but occaisionally there’s a subtlty in the question which means the seemingly-obvious answer is not the correct one. But may that was just me.
To prepare for the exam I used mix of Microsoft Learn’s free online resources, supplemented by access to Percipio resources through my work, and WhizLabs practice tests. As someone who has been neck-deep in databases for years, I feel I had a bit of a leg-up on some of the more general topics – which meant it was really just the finer points of the specific Azure data solutions I had to learn.
Part of the exam syllabus covered Business Intelligence reporting using Power BI, which might be a good starting point for the next exam in my calendar: Power Platform Fundamentals (PL-900)!
Since last year I’ve made a point that whenever I feel a bit “neurofunky”, I try to do something to invest in myself. The last few days have been a thing so I’ve planned the pathways to my next certification(s), and set myself up with some of the resources I’ll need to get there.
Right now, the plan is to complete the following exams over the next 6 weeks:
Azure Data Fundamentals (DP-900)
Power Platform Fundamentals (PL-900)
Azure AI Fundamentals (AI-900)
At least 2 of those topics are pretty much brand new to me, so it’s going to be an interesting time…
I’ve already booked the exams, to give myself a set timeframe and deadline for each. My employer offers vouchers towards taking these exams, so it doesn’t cost me anything over and above the time and effort investments. Even if the exams hadn’t been free, then I’d have probably booked at least one of these (or maybe more, just spread out over months rather than weeks)
So we’ll see how it goes. I’m excited to learn some new things, but I am conscious I’m going to be under a bit of stress due to the timing.
I finally decided it was time to upgrade my trusty TP-Link WiFi-N router to something a bit more modern. Even though it’s an older router, it’s been far more reliable and stable than just about any ISP-supplied router I’ve had the misfortune to use.
I finished setting up the fancy new “mesh” system in about 15 minutes. It was much easier than previous home networking gear I’ve used! Plug one node in, let the Linksys app do the work, then enter my ISP username and password. Once the first node is fully up and running, plug in the other nodes one by one and let the app do the rest. All the iOS devices “just worked” with the new network, apart from the HomePods – which needed to be “moved” using the Home app. Other than that: 2 LifX bulbs needed to be reset and readded, while the games consoles and Windows PCs needed the new password input (even though it was the same as on the old router).
WiFi connections seem to be rock solid throughout the house now, and browsing a few sites and services “feels” a bit snappier, but that could just be confirmation bias. There’s a parent node in the hallway, then child nodes in the busy areas – the living room and office – which ensure the whole house is well covered.
After installing a firmware update, the nodes all integrate with HomeKit, which is a nice to have. This lets me restrict HomeKit-compatible devices to only my local network, for security purposes. It’s not essential, but if it’s there, why not use it?
I guess the real test will come tomorrow when we’re both working and making Teams voice and video calls all day.
I got rid of my Hey email address last night. I really liked the “mental model” of how the app organised email, but there was never enough there to push me to fully switch over from Proton Mail. I couldn’t justify paying for 2 email services over the long term, so one had to go eventually. The furore and fallout of the last couple of days helped make the choice easier and sooner.
I’m not going to touch on the “no political discussion” topic, or the apparent executive power-grabs and other Bad Things which have come out recently. I have strong opinions on all that (the bits I’ve seen and read, at least), very little of it positive or flattering to Basecamp management, but it’s not what I’m writing about right now. Maybe later.
One of the less obvious details lurking in this seedy story is that Basecamp as a company seemingly can’t be trusted with customer data. That’s a dealbreaker for me. A list of “funny” customer names might seem innocuous at first glance, but it shows a petty disregard for your customers privacy, or who can access their information (never mind the basic disrespect of being made fun of). Every current and former Basecamp customer who has heard of the kerfuffle over the last few days will have probably asked themselves “was I on the list?”
If this list had been a short-lived “joke” by a bored employee abusing their access and privileges that would be bad enough, assuming management caught it in time and acted decisively in the customer interests. But the list lasted and grew for years. It was an institutional thing with many contributors over time. That in turn raises further concerns about oversight and tacit approval of a company culture which thinks it’s fine to abuse customer data and trust like this.
Apparently management didn’t know until someone raised concerns recently. That doesn’t pass the smell test for me. Something as apparently embedded into a workplace culture as this list isn’t magically invisible to The Boss, especially if it’s well known among the staff. They know what’s happening; they just know it’s better not to say anything so they can have “plausible deniability” when it inevitably and eventually goes too far. I’ve seen this sort of managerial behaviour too many times to believe any company – even one as supposedly enlightened as Basecamp – is immune.
But let’s suppose for a moment they were (as they say) completely ignorant – for years – of the list… what other abuses are they still unaware of? Are there other skeletons hidden in the closet? And how is that in any way supposed to inspire confidence in me, their customer, that they know what they’re doing to protect my data? Especially data as sensitive and far reaching as email?
No thank you. No amount of convenient email organisation is worth that. So: subscription cancelled, data exported, deleted1, and the app has been removed from my devices.
Has it been though? I guess that’s something I’ll need to take on trust… ↩