Vincent Parrett

@Eli M. Thanks. We're some way from that yet, but when we get to the IDE integration we'll be looking for people with UI skills to contribute. Right now I want to focus on the structure of packages/libraries etc, how they are defined, naming conventions, etc.

Chocolatey is Nuget repurposed for installing applications, not for installing code libraries. It's also a .net application. As much as I love .net/c#, I think a delphi package manager should be written in delphi.

I'm open to suggestions when it comes to package naming conventions, but it's going to take a lot to convince me to allow running of scripts. Running random scripts that haven't been vetted for security is major issue. Perhaps a better option would be to include some sort of dsl that has some pre-defined functionality (like copy files etc).

Uwe, I'm definitely open to contributions, there's no way I can achieve the desired result by myself (I still have to earn a living!). It's a complex task, and there are a lot of moving parts and design decisions to be made. I expected package unloading to be an issue. Package Magician looks interesting.. hopefully we can get you on board, I haven't really played with the open tools api much for many years. Github link fixed.

Thanks for the head up, I changed the project name at the last minute and forgot to update the link.

Expect layoffs and outsourcing to contractors.. standard Idera practice. Some interesting comments on hacker news.

Take a look at Continua CI and FinalBuilder for the best Delphi support in a CI/Build environment.

10.3 uses MSBuild 4.0 rather than MSBuild 3.5 in 10.2 - so make sure you specify that in the Continua CI MSBuild Action.

Hi All We have released FinalBuilder 8.0.0.2496 with support for Delphi and C++Builder 10.3 This page shows what's new/changed in this update : FinalBuilder version history You can download a fully functional 30 day trial version of FinalBuilder here. FinalBuilder is an award winning Automated Build Tool, with support for Delphi, C++Builder, Visual Studio and many other tools (over 600 individual actions). Regards Vincent Parrett VSoft Technologies https://www.finalbuilder.com

Not sure what's going on, just tried it again from work (was at home before) and it is loading a lot less on the second page load. It's a slow user experience though.

Just to provide a contrast performance wise, my forums which run on a self hosted (in LA) Discourse instance (ubuntu server 18.04 - 2 virtual cores, 2G ram) - granted I don't have anywhere near the the number of active users but the point is the forum software implementation : First page load Second Page Load (it feels almost instant). Ignore the poll request, that's just a websocket long polling request running in the background for notifications. If you look at the response headers for the requests you will see it marks the images etc as cachable.

Outback Hardly, I'm in Canberra, the capital city. I have a 95Mbit connection at home, 70Mbit at work. (both vdsl2). The problem is not my connection, it's the distance/latency. Pinging en.delphipraxis.net [138.201.18.17] with 32 bytes of data: Reply from 138.201.18.17: bytes=32 time=331ms TTL=52 Reply from 138.201.18.17: bytes=32 time=355ms TTL=51 Reply from 138.201.18.17: bytes=32 time=330ms TTL=52 Reply from 138.201.18.17: bytes=32 time=331ms TTL=52 Ping statistics for 138.201.18.17: Packets: Sent = 4, Received = 4, Lost = 0 (0% loss), Approximate round trip times in milli-seconds: Minimum = 330ms, Maximum = 355ms, Average = 336ms The issue isn't helped by the number of requests per page load : When latency is high it has a snowball effect on page loads. There also doesn't appear to be any caching of images, js or css (no caching response headers) for files that all have what appears to be cache busting filenames (so cache invalidation would not be a problem). There's a lot of scope for improving the performance, but sadly I can't move any closer to the server!

Kudos on getting this forum up and running, but please consider hosting it somewhere more central (like the US). From Australia each page load is painfully slow (5+ second page loads). Hosting in the US would even out response times for more users (obviously it would make it slower for europe).