AWF Slowing Down

[CJ_London]

I know of several Access forums that no longer exist.

So why not name them?

You seem to be offended by my non technical comments.

No

f Jon's not in it for the money, then why have Google adverts on AWF?

To recover his costs

 

[Edgar_]

@BlueSpruce
This forum stands a much better chance of survival by staying niche, rather than diversifying in that way. That's because each of those themes already has its own communities, which are often available at the cost of a free sign up.

I'd probably add a forum board to discuss strategies to make Microsoft open source the app, so the rest of the world can make the improvements it requires.

 

[Edgar_]

I'd probably add a forum board to discuss strategies to make Microsoft open source the app, so the rest of the world can make the improvements it requires.

Sort of like the recent "Stop killing videogames" iniciative, where players are imploring AAA game dev firms to stop shutting down the servers that games need, when they think the games aren't profitable anymore.

The initiative basically encourages these companies to open source their games, even in a controlled manner, allowing them to retain some oversight over what can and can't be done. That would enable players to continue enjoying the games long after their official shutdown. There are a lot of games that remain highly playable, with communities thriving decades after these games were released, such as King of Fighters, Age of Empires, GTA San Andreas, and I'm sure others here can name a few too.

Building on that idea, this forum could serve as a platform for other software packages facing a similar fate: programs whose developers are no longer willing to maintain them, yet whose user base is eager to keep them alive.

 

[Jon]

The short answer to all this is that since Google gives AI answers right in its search engine, these type of sites have seen their traffic plummet. It will lead to the demise of many sites as they become financially unsustainable. This then beggers the question, "Where will the bots get their future information from?" Perhaps it does not matter because by then they have already scooped up most of the data, who knows. The financial ecosystem of the entire internet is under threat.

The slow decline of Access is real, but the biggest impact lately has been Google's introduction of AI answers directly in search.

 

[Minty]

Buy out Google?

 

[CJ_London]

the AI answer is only as good as the question asked.

 

[Jon]

I don't believe Jon is in it for the money

What money?

 

[Jon]

To recover his costs

Correct. It is not free to run this site.

 

[Jon]

Thanks for that explanation. So what can be done to offset those threats?

I appreciate both your interest and concern for the success of the site. The non-Access areas were created to provide some kind of community for members to hang out. Over the next 5 years, I believe we will see such massive change both online and offline that we will wither on the vine. I wish it were different.

I will do my best to keep the site going so you guys can still keep in touch with one another, and get engaged in discussion, even if not Access related.

 

[The_Doc_Man]

the AI answer is only as good as the question asked.

IF you are lucky, because even with a decent question, there is also the factor of the rarity of the subject. After all, if there is very little training material in the internet, the LLM might not have run across it and thus not "know" much about it.

For instance, in modern terms, if you now ask for the meaning of the word "parachor" you will get a physical-chemistry answer having to do with surface tension in a liquid where the surface is exposed to open air (and thus, atmospheric pressure.) You measure it based on the meniscus of the exposed liquid. It is indirectly related to the compressibility of the liquid, and therefore is of interest wherever you have hydraulic lines or anything else transported by tubing.

Back in the 1960s it had another meaning. In the days when virtual demand-paging operating systems were coming into focus, the "parachor" was a task-specific measure of program efficiency as a function of how much physical memory was allocated to that task. There is an inflection point in the graph of efficiency vs. allocated memory. To the "less memory" side of that inflection point, you get dramatic increases in efficiency as you add memory. To the "more memory" side of that point, adding more memory has a far lesser effect. That inflection point is the parachor for that task.

The "rarity" comes into play because these days, it is possible to get 16 Gb of RAM pretty easily, and despite the nearly innumerable service tasks running in Windows (run Task Manager, Processes display to see them all), you usually have over half of physical memory still available. However, in the days of 4 Gb of RAM being considered a huge machine, a LOT of stuff ran as virtual processes and only paged in when needed. In that context, the parachor was a subject of much discussion. Now, not so much.

 

[AccessBlaster]

The issue, as I see it, is that people are getting better at prompting AI, and with every bad prompt, the learning curve gets less steep.

 

[The_Doc_Man]

So then the bot should reply with "There's more than one meaning for that word. Please select from the following meanings...". Wikipedia does this whenever there's ambiguation.

While I agree with you (in reference to Wikipedia's disambiguation pages), that didn't happen for my search. Which was the point of the post. Some things are researched so rarely that their "hit" rate dooms them to obscurity as LLMs and their prioritization algorithms take over the task of simple web-searching.

With the advent of 8 GB and 16 GB memory cards, the problems measured via the parachor are now mostly "out of sight, out of mind." When you have 32-bit memory-model tasks swimming in a 32 GB pool of memory, there is usually very little to stop Windows from loading the whole task into memory. When I look at Task Manager's Processes page, it shows me memory usage, and all but 4 of the individual tasks I see are <100 MB-sized, no tasks anywhere NEAR GB-sized. And so the issues with page management are a long-forgotten footnote in the depths of the "great Google mind." The parachor (disambiguation: computer memory) has been relegated to the dustpan of history.

 

[The_Doc_Man]

It is forever ensconced in the IBM Journal of Research and Development from the late 1950s/early 1960s, which is where I first learned about it. But I have a feeling that these days, most LLMs look for more modern references to pirate ('scuse me... learn) technical data.

 

[Isaac]

Convince Microsoft to actively market Access as a useful tool and a complement to SQL Server. The SQL Server team is convinced that "Access" which they can't separate from Jet/ACE is their competitor. There really isn't anything Jon can do to draw views. A lot of the decline can be directly attributed to changes in the way Google works and how it ranks pages. 20 years ago, I could google myself and I would show up at the top of the first page due partly to my posts here and partly to my high finishes in Bridge tournaments. Today, I am a non-entity.

I'm surprised you ever came in that high on a google search, in a country with 300 MM people and the name Pat Hartman, too common to result in google searches. I on the other hand am the ONLY person who comes up when searching for my name. that's the curse of having an unusual name - easy to find.

 

[Isaac]

Wrong, there's plenty alternatives, e.g. DotNet, C#, Web Apps, AI, etc. The proof is virtually no one wants to learn to develop with Access and stakeholders don't want any new apps developed with it. It's only relevant to our Cottage Industry and the users who have been clinging onto it. How long is that going to last? It's dying a slow death by attrition.

But replacements have always been there. It's not a trajectory, it's a static situation.
And the fact of people creating new platforms every time they turn around, that's been that way a long time too.
Nothing effective can really replace Access for desktops on a LAN or with Citrix or any other manner of virtual servers.

However, you're smart....starting an "Is Access Dead?" conversation IS a proven way to attract posts, since that's what you're gunning for.
Be aware, you're likely to be labelled a troll, as that's usually the convo that trolls start.
Got anything to say about graphics on forms?

 

[The_Doc_Man]

If AI replaces humans everywhere, then what purpose will we have?

Without us, who would grumble about poor service and inadequate phone menu systems?

 

[AccessBlaster]

then what purpose will we have?

Copper tops?

 

[The_Doc_Man]

Scary thought that we could be used as slaves to produce more energy for powering hungry AI datacenters and their bots. AI would also conclude its not worth saving us from ourselves and the world is better off without humans who are inferior to them, pollute, spread disease, and create wars.

Which is why we should NEVER build those future AI computers that don't have electrical cords. We must never forget that old adage that tells us "A computer's attention span is never longer than its extension cord."

 

[AccessBlaster]

The enormous amount of energy AI will require in the future may be its downfall. I can envision massive power plants running on fossil fuels and coal just to keep up with peak demand from all things electric. This will not be welcomed by certain climate groups.

 

[The_Doc_Man]

The enormous amount of energy AI will require in the future may be its downfall. I can envision massive power plants running on fossil fuels and coal just to keep up with peak demand from all things electric. This will not be welcomed by certain climate groups.

An AI center is being built in Louisiana, near the city of Alexandria (center of the state). People all around the state have already started screaming about how the implied increase in demand is going to futz around with the supply/demand equations and raise utility prices for the average person who probably would have little to do with AI. My thought is that if the AI company can't afford to build their own self-sufficient, independent source of power, they can't afford to build the AI either.