|
I daily travel by train, for the reason it safer than driving on the roads around Bristol with the Mad Max wannabes driving...
|
|
|
|
|
Just outfit your car with metal horns and a couple of chainsaws, and join 'em!
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Today is not a good day too die (it's Friday! AKA POETS day)...
|
|
|
|
|
Amusing story - and very Japanese...
I travel by train quite a lot - being a non-driver - and have to say I quite like it, and don't have the stereotypical view of British train travel that many do. In my experience, they usually run more or less on time, and the staff generally pleasant. They could often do with more seats though, and while the toilets are sometimes not as clean as you'd like, the fault for that must surely rest primarily with the passengers... sorry tot say it, but the British are not a clean people. Just walk around London - it must be the filthiest capital city in Europe by some margin. But generally, I like trains!
|
|
|
|
|
Quote: They could often do with more seats though, Too true I object to paying for my season ticket and having to stand.
|
|
|
|
|
I use a lot of trains, too - I don't drive either.
I think the experience varies a lot from region to region but to my mind, the major issue these days is over-crowding rather than punctuality.
This is something that could be very easily cured. The reason for it is that Railtrack (or whatever they're called this week) rent their lines on a cost per carriage rather than a cost per train basis. A simple switch on that policy would make life a whole lot better for many passengers.
Sometimes, though, you really have to question the operators - London Midland choosing to use a 3 car set on the 5.15 departure from New Street on the main commuter line in Birmingham, for example, rather beggars belief when there are 6 car sets operating at non-peak times.
98.4% of statistics are made up on the spot.
|
|
|
|
|
[begin diary]
I'm now the owner (proud or not) of a new Note8, having dropped my ancient Samsung 5 hard on the floor yesterday night, resulting in the screen displaying randomly colored pixels. I guess I have a new random number generator, haha.
Flushing the toilet this morning, there was a strange gurgling noise. Check the basement, yup, the septic line is backed up. Again. Won't get snaked until tomorrow. I should really invest in one of those motorized snakes, this is a regular enough occurrence that the investment will pay off in a couple of years.
On a positive note, I've been fussing with .NET Core 2.0 and one of the core pieces of my common library for building things compiled and my initial tests indicate everything is working fine, no code changes. .NET Core 2.0 is definitely an improvement over 1.0!
And while I ditched Docker for Windows, the Docker Toolbox (that sets up an Oracle VM) is working great, I have bidirectional HTTP communication between the host and the container working without much of a fuss. Article hopefully before the end of the year!
[end diary]
|
|
|
|
|
Marc Clifton wrote: Flushing the toilet this morning, there was a strange gurgling noise. Check the basement, yup, the septic line is backed up. Again.
For me, its my kids. They haven't quite mastered the proper way to use the toilet without making it go boom. Any more info, and it would be TMI.
|
|
|
|
|
Marc Clifton wrote: Flushing the toilet this morning, there was a strange gurgling noise
[...]
Marc Clifton wrote: .NET Core 2.0
I thought for sure this was going to lead to a rant linking these two items together.
For once, I was disappointed.
|
|
|
|
|
dandy72 wrote: For once, I was disappointed.
Sorry about that. I'll "aim to please" next time.
|
|
|
|
|
Well...I'm still taking your praise for .NET Core 2.0 as a good thing. So I'm not entirely disappointed (especially since I'll have to start using it at one point or another).
|
|
|
|
|
dandy72 wrote: I'm still taking your praise for .NET Core 2.0 as a good thing.
So far so good, though at the moment, I'm trying to get NuGet (which I despise) to install Newtonsoft.Json, which is supposed to be core 2.0 compatible, as it's part of the ASP.NET package for core 2.0. But I'm not using asp.net, so we'll see what happens.
I wish I could just download and reference the DLL. But I saved a copy of the whole project before going down the NuGet route, so whatever it ends up installing, I should be able to add as a dependency.
BTW, one thing I find "interesting" is that the csproj, which is now an XML file, doesn't reference any of the source code files in the project, it seems that dotnet build simply picks up every .cs file in the folder AND sub-folders (which is what I want.) I guess they've tried to make it as brain dead simple as possible for all those Linux devs.
And I must say, it's really cool debugging the application in VS2017, working out the kinks, and then deploying it to docker.
[edit]NuGet stayed stuck on "Installing Package" so I finally gave up, cloned the source, compiled it, and referenced the .NET Standard 2.0 DLL directly, now my Json serialization works! [/edit]
modified 16-Nov-17 17:44pm.
|
|
|
|
|
Marc Clifton wrote: BTW, one thing I find "interesting" is that the csproj, which is now an XML file, doesn't reference any of the source code files in the project, it seems that dotnet build simply picks up every .cs file in the folder AND sub-folders (which is what I want.) I guess they've tried to make it as brain dead simple as possible for all those Linux devs. So it treats all Core projects like it's opening a website directly in VS? Go figure.
Jeremy Falcon
|
|
|
|
|
Jeremy Falcon wrote: So it treats all Core projects like it's opening a website directly in VS?
Who knows.
One amusing thing though is, I tried to create a sub-sub-folder in VS2017 (in a .net core 2.0 project) and while it created it, I can't change the name. The sub-sub-folder's name is always "New Folder". There's no rename on its properties, I can't even delete it. I had to exclude it and delete it in the file explorer.
|
|
|
|
|
|
Jeremy Falcon wrote: So do you feel your productivity has increased?
Productivity falls under Einstein's laws of relativity. In other words, don't ask my gf.
|
|
|
|
|
|
I got so pissed with nuget I now have a project dedicated to downloading packages, I then copy the dll's to the actual project for reference.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Mycroft Holmes wrote: I got so pissed with nuget I now have a project dedicated to downloading packages, I then copy the dll's to the actual project for reference.
Exactly! SOP here too.
|
|
|
|
|
Marc Clifton wrote: .NET Core 2.0 is definitely an improvement over 1.0! How do you like Core in general? Admittedly I'm a total n00b to it, but I'm curious to know your thoughts about your general impression of it.
Jeremy Falcon
|
|
|
|
|
Jeremy Falcon wrote: How do you like Core in general?
So far I'm impressed. It's a major improvement over core 1.0, in that one of the things I do a lot of, reflection, is working quite nicely. Other stuff we've come to love like Linq is working great too.
One annoying thing, when I add a reference to the "dependencies" graph, it doesn't set the "copy local" to true. In fact, it's blank. So the DLL doesn't show up in the bin\debug\netcoreapp2.0 folder, leading to a "unable to load assembly" error. Manually setting "Copy Local" to true fixed the problem, but still, Microsoft is dropping some bits, given my bizarre sub-sub-folder problem I mentioned in my other response.
What I'm trying to achieve ultimately should be fairly interesting but arcane -- I want to create a website where people can publish (as source code) microservices, like "verify the address with the USPS" or "get me the weather forecast" -- basically a library of public (and private) microservices from which you can build an application that get triggered automatically when a particular piece of "semantic data" is available. The output needs to be in some DSL that I'll have to create that will generate the HTML snippet to insert into the web page so you can view the results. The other fun part is slopping the code into a Docker container, building it there and running it in the container, so people don't blow away the server itself. I figure I'm opening a can of worms of malicious evil-doers, so at some point I'll have to figure out how to block access in the VM to the outside world, and only "approved" coders can write microservices that hit the web.
Anyways, that's the plan. It's a web version of what I've been working on with HOPE and FlowSharpCode, etc., and with .net core 2.0 I think I can finally support C# and true semantic types, rather than kludging it in a language like Python.
Sorry, probably more than you wanted to know. On the other hand, if you have some free time...
|
|
|
|
|
Ya know Marc, I see a pattern here. You're a bit of a forward thinker. I recall the MyXAML days in particular, when most people didn't know about declarative programming. This project sounds cool, considering the "semantic web" is a fundamental building block to Web 3.0 concepts. I'm not really sure what DSL means in this context, but if what you're getting at is self-building services that get served up depending on logic then whether I get it or not, I do know that's where we're headed in the future with AI and computers programming themselves. Bye bye everything being so static ya know.
You gonna write an article about it so us mortals can read about it?
Jeremy Falcon
|
|
|
|
|
Only a bit.
Take a look at this image: weather forecast Sorry it's rather small, I posted into my WordPress site. It was created by hitting the Weather Underground API for the weather given the semantics for an address (WU uses the city and state). The screenshot is a WinForm windows generated on the fly. All this stuff is already running as microservices on the desktop.
But for the web, take the same code and the DSL would specify how the content would be rendered. I don't want people having to use the crap that we know as HTML/CSS (unless they know what they're doing, like you) but rather, take the semantics of the output and define how, on a grid, you'd want the information layed out -- text, images, etc. So the DSL is a user-friendly way of defining where content goes, and the engine converts it into HTML.
|
|
|
|
|
Ah. I got ya. Kinda reminds me of MyXAML then in principle. Bootstrap works in a similar way actually and once you get used to grid-based structures it works pretty well.
As an aside... Just as long as we call them grids and not tables. That’s a dirty word despite it being a similar concept. Bad bad tables. But grids are cool. Go figure.
Anyway, I like the way your mind works man. Except when you dis JavaScript. The project is an interesting concept.
Jeremy Falcon
|
|
|
|
|
So, whilst my brains leak joyously out my nostrals and into soft pieces of paper, I remembered a question that I thought to pose before this ill-ustrious forum:
Why are credit scored ranged from 300-850 ? I mean, really, you could be dead and still have a 300 score. Why not start at zero? For that matter, one could also argue how there can be a cap on credit scores. Would a billionaire who always paid their debts, never declared bankruptcy, etc. really be limited in their credit worthiness (as if they needed credit).
Let's forget the latter question: why not use 0-550 for credit scores? A range of 550, itself, is rather odd, too.
So who knows what and why - or make up some stuff.
Ravings en masse^ |
---|
"The difference between genius and stupidity is that genius has its limits." - Albert Einstein | "If you are searching for perfection in others, then you seek disappointment. If you are seek perfection in yourself, then you will find failure." - Balboos HaGadol Mar 2010 |
|
|
|
|
|