March 15, 2025 at 12:00 am
Comments posted to this topic are about the item Widespread New Technology Adoption
March 15, 2025 at 4:33 pm
The downside of adopting new technology is that each year there is new technology and what was new last year is now old. So if you by default choose the latest, you end up creating an awful mess.
Just look at the first ASP.NET applications, now 24 years old. You can see the layers of technology, coming and going, like a geologist looks at sediments. ("Oh look, somebody tried Linq To Sql ... and then abandoned it!") And yes, that new stuff makes it into products and stays there when it's age-old.
March 15, 2025 at 4:35 pm
The worst is Microsoft's data access. After it had had DAO, and then ADO, it started with ADO.NET in 2021. A year after that the Data Access Application Block (DAAB) was introduced, and justified because it embodied "Best Practices". When the Microsoft team introduced it I had the audacity to ask them: "Why didn't you put these best practices in ADO.NET itself?" which was followed by an embarrassingly long silence. After this, we got DAAP 2, which had a completely different interface. (So many new best practices in just one year!), followed by DAAP 3, DAAP 4, DAAP 5, etc. Then we got some Linq To Sql, and then EF, EF 2, EF 3, EF 4, EF5, EF6. After EF4 came out, the product manager of EF Julie Lerman complained in Visual Studio this was her third complete rewrite. Did they finally get a stable product? Nope. EF Core 1 was not first time right, and was followed by EF Core 2, EF Core 3, ... you get it.
I was particularly annoyed when I had to fix a bug in code using DAAB 1, when Microsoft had already taken down all documentation because it thought nobody was using it anymore. Luckily the internet has a long memory (thanks Code Project!).
A final remark: Yes, doctors are eager to prescribe the latest pill. That's why they don't heal anymore and so many people go sick through life, even with record health insurance costs. Analyzing the root of a problem leads to the simplest and most effective solution. The latest may make you feel good (like in a brand new Tesla) but it is hardly ever the greatest.
March 16, 2025 at 3:08 pm
Steve, I used to think that investing in myself would result in others following suit. Sort of rising water lifts all boats.
However, I work for an organization that is dead set against investing in its employees. Upper management can go to annual conferences, etc., but the common working man or woman haven't been to any training or conference in over 30 years. I've been advocating for training on GitHub Actions, especially with self-host runners, for over a year. I've finally gotten enough interest so that upper management has agreed to that, but I'm also seeing a lot of resistance against it. One of the upper managers is trying her best to get the training done on the cheap, by having a contractor do it for free. I've never worked at a place which does its best to keep everyone's skills at very old levels such as where I work. I do train on my own, paying for an annual Pluralsight subscription, but what really bothers me is how upper management will NOT allow me to use anything I learn. We all must work with technologies that are at best 15 years old. What's the point of learning anything new if you cannot use it? I have some hope that continuing to learn something new may result in my getting a job where I can use it, but it doesn't help not being able to practice what I learn and given the current job market, I feel pretty hopeless.
Kindest Regards, Rod Connect with me on LinkedIn.
March 17, 2025 at 3:43 pm
Fun article, as I'm passionate about EVs and enjoy learning new things. However, the reason I don't have an EV right now is because there aren't any on the market that I want. Even back when I had Tesla stock and didn't hate Musk with a passion, I didn't want a Tesla for a variety of reasons. I understand that we are on the verge of lots of new electric vehicles options coming in the next few years though (assuming the current administration doesn't mess that up), and I'm super excited for those opportunities.
All of which is to echo some of the previous comments that sometimes we don't adopt new tech for the reason that it isn't right for us or we feel that it isn't developed enough yet to fully adopt in a production setting. (Though I feel Rod's pain! "Ugh" for his management.) I feel that too many people rush into the latest thing without carefully evaluating it. There's what I think of as a success bias (maybe there's a better term?) with new tech: People see the pluses of the new and minuses of the old while ignoring or downplay the minuses of the new and the pluses of the old. This failure to do a proper analysis of the net benefit/detriment is the bigger problem in my opinion.
March 17, 2025 at 3:57 pm
I also want to say that I really appreciate people who bought Teslas years ago, as I believe that the early adopters paved the way for EVs to be popular in the US now.
Also, my friend recently purchased a plug-in hybrid. Her plug-in hybrid operates on all-electricity for all her daily, in-city driving. If she needs to take a trip to another city, going past the battery range, the gas kicks in. She plugs the car into the regular garage outlet at night, needing no special plug-in. I'm committed to having my next vehicle be a full EV (if I can find one to my specifications). However, if there were more plug-in hybrid models available to the US public and at an affordable price, I think a lot more people would adopt them. The plug-in hybrids get rid of that range anxiety, while greatly minimizing the need to get into lines at the gas station.
Comparing that situation to AI: I think the way forward for AI will also be a hybrid situation at first, much as explained by Steve's article. Trusting AI too much is not a good thing at this point. Ignoring AI completely probably has more harms than good. Finding the ways that AI can help, while attempting to minimize the negatives, is probably the way forward in my opinion.
March 22, 2025 at 11:28 pm
Steve, I used to think that investing in myself would result in others following suit. Sort of rising water lifts all boats.
...
This does happen, but there are always places that are exceptions and don't want to change. If the culture is so entrenched that it won't move then you can always move on.
It can be tough to find a job now, but showcase what you have learned with new tech. Don't depend on a resume. Blog, make a public project, showcase skills.
March 22, 2025 at 11:28 pm
Steve, I used to think that investing in myself would result in others following suit. Sort of rising water lifts all boats.
...
This does happen, but there are always places that are exceptions and don't want to change. If the culture is so entrenched that it won't move then you can always move on.
It can be tough to find a job now, but showcase what you have learned with new tech. Don't depend on a resume. Blog, make a public project, showcase skills.
March 22, 2025 at 11:30 pm
Fun article, as I'm passionate about EVs and enjoy learning new things. However, the reason I don't have an EV right now is because there aren't any on the market that I want. Even back when I had Tesla stock and didn't hate Musk with a passion, I didn't want a Tesla for a variety of reasons. I understand that we are on the verge of lots of new electric vehicles options coming in the next few years though (assuming the current administration doesn't mess that up), and I'm super excited for those opportunities.
All of which is to echo some of the previous comments that sometimes we don't adopt new tech for the reason that it isn't right for us or we feel that it isn't developed enough yet to fully adopt in a production setting. (Though I feel Rod's pain! "Ugh" for his management.) I feel that too many people rush into the latest thing without carefully evaluating it. There's what I think of as a success bias (maybe there's a better term?) with new tech: People see the pluses of the new and minuses of the old while ignoring or downplay the minuses of the new and the pluses of the old. This failure to do a proper analysis of the net benefit/detriment is the bigger problem in my opinion.
On the second part, agree. People rush into tech without really evaluating it well.
On EVs, I'd be wary of new model/new brands as the software part of an EV is hard. I think Ford took a few years to get it right with the MachE. Tesla certainly took quite a few years. Hyundai/Kia seem to be doing well and I'd look at one of those Ionics if I needed a new one.
March 22, 2025 at 11:31 pm
I also want to say that I really appreciate people who bought Teslas years ago, as I believe that the early adopters paved the way for EVs to be popular in the US now.
Also, my friend recently purchased a plug-in hybrid. Her plug-in hybrid operates on all-electricity for all her daily, in-city driving. If she needs to take a trip to another city, going past the battery range, the gas kicks in. She plugs the car into the regular garage outlet at night, needing no special plug-in. I'm committed to having my next vehicle be a full EV (if I can find one to my specifications). However, if there were more plug-in hybrid models available to the US public and at an affordable price, I think a lot more people would adopt them. The plug-in hybrids get rid of that range anxiety, while greatly minimizing the need to get into lines at the gas station.
Comparing that situation to AI: I think the way forward for AI will also be a hybrid situation at first, much as explained by Steve's article. Trusting AI too much is not a good thing at this point. Ignoring AI completely probably has more harms than good. Finding the ways that AI can help, while attempting to minimize the negatives, is probably the way forward in my opinion.
We've had a few Prius's and loved them. I'd certainly consider one again if I needed a bunch of driving on many days. As it is, I rarely go over 200mi/day and the Tesla does fine. There are times we do and either we plan for a 15-20min charge somewhere, or we take another car. And yes, I am lucky we can have two cars
March 23, 2025 at 1:48 pm
Rod at work wrote:We all must work with technologies that are at best 15 years old.
Would upper management actually realize if you used .Net 8 instead of .Net 2? Would they realize if you asked for a test vm and put the Data API builder on it? etc
Upper management wouldn't recognize whether we're using .NET 8 or .NET 2. They probably know recognize the delay in development, though.
Kindest Regards, Rod Connect with me on LinkedIn.
March 23, 2025 at 5:54 pm
Ken McKelvey wrote:Rod at work wrote:We all must work with technologies that are at best 15 years old.
Would upper management actually realize if you used .Net 8 instead of .Net 2? Would they realize if you asked for a test vm and put the Data API builder on it? etc
Upper management wouldn't recognize whether we're using .NET 8 or .NET 2. They probably know recognize the delay in development, though.
Just start slipping new stuff in. Start with small projects you are doing by yourself. If anyone asks just say you are aware of the conservative approach but did not know you actively had to use old code with proven security vulnerabilities. (Any old code is bound to have security vulnerabilities!) Things like the Data API builder might actually speed up development.
You obviously work for a bureaucratic organization. People get promoted in bureaucracies by not making mistakes; they do not have to achieve results! The only way not to make a mistake is not to make a decision. This is why bureaucracies have lots of committees, and sub-committees, so all decisions are spread among a number of people and no one can be blamed if things go wrong. When things go well members of the committee take the credit and are more likely to be promoted. If you have modern code working well in production then no one is likely to make a decision to remove it. In fact a member of upper management might take the credit which might even mean you are more likely to get a promotion. ie Start modernizing the code in areas you think will benefit the most as you are the one taking the slight risk.
March 23, 2025 at 9:41 pm
I like your reasoning, Ken. You are correct, I work in a highly bureaucratic organization. I work in State Government. My previous boss, who retired last year, was one of those who ordered me to leave everything alone. His argument was all developers hated all other developers work and wanted to rewrite everything. I didn't agree with him, but also knew I couldn't win that argument. It was strange that he was so adamantly opposed to at least upgrading the frameworks involved, as he was just a first line manager. Well, I'll see what I can do.
Kindest Regards, Rod Connect with me on LinkedIn.
March 24, 2025 at 10:21 am
It was strange that he was so adamantly opposed to at least upgrading the frameworks involved
Not really. If most people are happy with the way a system is working why risk any change. As far as your manager was concerned there were no upsides. Even a sucessful update will have shakedown problems and everything will just seem to work as before.
If you want to update an old system with personal data the best approach is to push security. No bureaucrat will want to be blamed for loosing voter details and network security can no longer be relied upon. Maybe:
1. Work out what you want to do.
2. Push for a security audit.
3. Get the system on the risk register. (Bureaucrats love risk registers.)
Once on the risk register you should eventually be able to get time and funding for the upgrade. Also, following best security practises might lower insurance premiums. I think .Net 4.6.2 goes out of support in January 2027 and .Net 3.5 SP1 goes out of support in January 2029. Older versions of .Net are already out of support so are potentially vunerable.
Viewing 15 posts - 1 through 15 (of 16 total)
You must be logged in to reply to this topic. Login to reply