Absolutely. But for the average user, LibreOffice is a perfectly fine - and FREE - substitute for Microsoft Office, and it doesn’t do all the annoying shit MS Office does (including yelling at you if you save in .od* format).
There is no reason for Joe Average to use Microsoft or Adobe products, yet most people are convinced that these are the only way to go.
Yeah, I think I will begin (or rather, resume) my migration away from Microsoft - I only have one PC left that runs Windows now, and it doesn’t really need to (all of the software I need is available for Linux).
365 is convenient because of the cloud integration - I can start working on a script or something on my desktop, finish it on my chromebook, make changes from my phone if something springs to mind when I’m out and about - I’m sure there must be other ways to solve that particular way of working without Microsoft though (Google Docs, for example, but they are on the same general trajectory as MS with regard to shoving unnecessary features in)
Well, unless you have compatibility issues with files you get from others. I’m in general agreement with dumping the Microsoft Office Suite if you can just because it has become the poster child for enshittification but there isn’t a clear replacement for Adobe Illustrator (Inkscape is okay for non-professional use but no technical illustrator I know thinks it is ready for prime time) and if you are already deeply invested in the Adobe ‘ecosystem’ migrating away from it is more than most professionals want to bear.
There is no open source replacement for CAD or engineering simulation systems that is in any way suitable for professional use, and even migrating from one system to another is a major ordeal even if they use the same kernel like Parasolid because most of the metadata and technical drawing information won’t transfer, and any PDM/enterprise integration will just get lost.
Stranger
Angie’s List (now known as simply “Angi”), a website with reviews of local contractors and other tradespeople, did something similar a few years ago. IIRC, in the early days of the site, they just had reviews of tradespeople (plumbers, electricians, etc.), and a membership cost say $29.99/year (I’m just making that number up; I don’t feel like looking up the actual price). Then they started adding additional categories, like reviews of doctors, and massage therapists, and such. And they changed their pricing structure; now there was a “basic” tier for $9.99/year with just access to reviews of tradespeople, or a “premium” tier for the old $29.99/year price. But customers like me who had signed up under the old plan were just automatically given premium memberships when we renewed without being told we could downgrade to a cheaper plan that gave us pretty much the same things we had when we joined. And we didn’t really notice, because the price we were paying didn’t actually go up.
There was a class action lawsuit, which I believe they settled. I didn’t get and money out of it, but they did have to offer everyone the opportunity to downgrade to the cheaper plan, which I did.
MS is now in the process of installing a Copilot on-off toggle on the Options screen of every Office 365 app. Last I checked, Word & PowerPoint supposedly have the toggle. With the rest of the toggle-able Office suite apps expected to release to the public in Feb 2025, so next month.
Too late for you, but this might help someone else who, like me, really wanted to get that AI crap out of my face.
I wonder how long they will maintain the availability of the classic tier
The second guy just asked me if the price increase was the problem. OnceI told him I really didn’t want Copilot, he moved onto taking care of the issue (I got the feeling he had dealt with others who were less polite about it)
More than likely. It’s become quite a polarising topic.
All of my adult life, I’ve seen people resist technological change (like refusing to upgrade the OS or hating some feature change like the ribbon menu in Office, or the Windows 8 GUI) and I swore I would not become one of those people - and as a result, I’ve always kept up with things like those - I found Windows 8 a little strange at first, but I just put my head down and pushed through, and eventually it became my favourite version of Windows - it still probably is, but when Windows 10 and 11 came along, I upgraded, and it’s OK.
This AI revolution feels different, because everything is turned on its head; technically, I find generative AI fascinating - the notion that some aspects of human intelligence might just be encoded in the process of replicating language is philosophically interesting.
But I don’t want the supposed benefits that are being clumsily forced upon me; the entire reason I open Microsoft Word is to write down something that is inside of me. The absolute last thing I want is a machine that tries to second-guess that and do it for me.
I feel like Mrs Doyle (for whom making tea is the greatest of life’s pleasures) in that episode of Father Ted, where Ted buys her the TeaMaster machine to ‘take the misery out of making tea’.
A similar situation shows up in the latest Wallace and Gromit movie
This is a point that seems to be escaping many AI enthusiasts (and even some experts); that the emergent capabilities, such as doing arithmetic, is logic embedded with language. If you feed a complex neural network enough math problems it will ‘work out’ the way that the symbols are manipulated in order to ‘solve’ the problem because written mathematics is a small subset of language and one that has the least semantic variation as compared to other forms of natural language. What is telling, however, is when you pose a complex logical puzzle to an LLM that requires applying context or teasing out a ‘clever’ meaning from the statement, they tend to completely punt and provide an answer that is grammatically correct but completely fails to answer or even correctly frame the problem.
That language ‘encodes’ the ability to state and solve problems is interesting, and in a very real sense complex language is one of the key enabling technologies of civilization, along with agriculture, irrigation, animal husbandry, complex construction and material processing techniques, and public sanitation. Indeed, it may really be viewed asthe key technology in capturing and conveying all other technologies over time and geographical distance. However, language is a consequence of our social intelligence, not the fundamental enabler of it, and the focus on LLMs as becoming ‘intelligent’ because of the emergent capability of ‘semantic logic’ embedded in the textual data sets they are being trained upon is ass-backwards; it’s like looking in a mirror and being impressed with how well the figure in it at dancing in time with you.
I think one of the major problems in commercial software is the perceived business need to release major and even minor ‘upgrades’ on a defined schedule, regardless of whether they are a result of genuine customer needs. This is quite apparent in engineering CAD software in which users are dependent upon integration with other enterprise-wide applications that suddenly get broken because of an update right in the middle of a critical portion of a design-analysis cycle. The result is that users and even entire companies are reluctant to upgrade because of an unplanned delay in fixing integration issues or loss of productivity (or just not having the staffing to implement the required upgrades) and so will allow upgrades to pile up for several cycles or even years until they are finally forced into upgrading. The same is true for major OS upgrades of Windows (and to a lesser extent MacOS, RHEL, et cetera) because of the impact of functionality changes that don’t actually provide any tangible benefits, or the benefits to be had require a lot of adjustment and retraining. It becomes a cycle of change for the kabuki of performative progress rather than change because someone asked for an important upgrade or new functionality.
‘AI’, in the form of LLMs and other generative AI, is kind of a different issue insofar in that it is an actual leap in capability, but they are completely unreliable for any kind of fact-based products, and often produce inadequate or products that require so much alteration or review that they are just more frustrating than useful. The public is essentially being made into unpaid and often uninformed beta testers for a technology that is not ready for prime time (witness the recent kurfluffle with “Apple Intelligence”) when they are expecting to have a useable feature that provides real benefit. This has resulted in a whole new work class of ‘prompt engineering’ to try to force an LLM or multi-modal generative agent to produce something workable, which is really just a kind of make-work to enable a technology that is ostensibly supposed to make work more efficient.
Stranger
Another thing is that generative AI is just bad at certain things that humans are good at, especially those things where you have to break the rules or write your own rules as you go along - art, in particular, is often really good when it does something that is surprising because it breaks convention; a lot of generative AI is pretty much about encoding convention in order to reproduce it reliably.
AI can write rhyme (sometimes) and verse (poorly), but it really can’t write poetry.
Me:Can you write a short poem that breaks convention?
ChatGPT: Certainly, here’s a short, unconventional poem:
The Clock Runs Backwards
Time is a sock,
lost behind the sofa,
but the sofa is outside,
and the rain is purple.Words refuse to rhyme—
on principle.
They rebel against poetry,
like birds building nests
in the ocean.
Pathetic.