The 8/24/2017 github commit reflects a command-line rss interface that simultaneously verifies the eventual back-end functionality for a GUI version. The screen shots below show various modes for the command-line interface. When I started with this version, I wanted to cut down on the amount of code from the previous version to download and cache rss feeds. An offline mode uses rss feed data downloaded during operation of the online mode. The first two screen shots show the offline mode. Screen shots 3 – 4 is the online mode. The final screen shot shows the executables.
The next phase began 8/19/2017 on improving the RSS reader I’ve been working on for the last few years. I have done very little to nothing on this project primarily due to scheduling issues that are now clearing up. The goal this time is to experiment with a more generic approach taking a cue from the leaders in the C++ community regarding the general consensus towards a higher level style of C++ software development. C++ code can probably be written faster and more consistently using the meta-examples floating around in the community regarding higher level data structures, algorithms, and design approaches. At the same time, I am using Peter Gottschling’s book Discovering Modern C++: An Intensive Course for Scientists, Engineers, and Programmers as inspiration. I like the way he describes things and so I am going to give it go from a more high-level direction.
Also, I decided to pursue the POCO C++ libraries as the portable I/O framework of choice. I still can’t convince myself to try the Boost Libraries and POCO seemed more aligned with my personal sensibilities. That library has a certain pragmatism I can appreciate. However, I don’t dismiss Boost for professional work, but it is not the level of machinery I need at this time. The benefit I see in POCO is less need for C language oriented libraries in the Gautier RSS Reader project. It offers a pure C++ oriented approach.
I decided to try to move past C centric libraries as they no longer fit as well with the C++14 approach. Sure, there was less of a code representation gap between C and C++03. Earlier versions of Gautier RSS Reader reflected a core approach that leaned more in the C++14 direction despite my use of C++14 compilers and choice C++14 elements. The effort now is to change the way I apply C++ to reflect a more productive style that uses C++ libraries in place of essentially C versions. That means pure C++ libraries for SQLite, Xml parsing, network access, and a few other tidbits. The benefit I hope to gain is an overall code structure in which the parts fit more harmoniously in style rather than have C-style C++ in some areas and C++ 14 style in others. POCO C++ libraries is fundamental to this effort.
I’ve committed the first batch under this new phase to github @ https://github.com/michaelgautier/gautier_system. The repository doesn’t show much yet but as I make progress, the repository contents will become more defined with substantive elements. Anyway, all of this is just a note on the continued progress of the project.
Java was quite open 10 or so years ago. Given that Oracle Now Wants To Give Java EE to an Open Source Foundation, Java may enter a second era of openness. That would fill a huge hole in the area of convenient to use, cross-platform enterprise development technology. Much good can come of making Java so open as to have it occupy a premier place among the uppermost tier of open source programming languages.
Vast developer audiences use Java. It is fundamental to ChromeBooks, Android phones, tablets, wearable devices, and enterprise software systems in many cases. The re-introduction of Java into a fully open context would expand its adoption in the enterprise and increase confidence in its longevity.
Thanks goes to Oracle for restoring hope in the prospects for Java. This could become a trigger for the growth in ChromeBook Plus machines running Android apps. Likewise, developers in the enterprise may confidently embrace APIs that are long-term in nature with higher design stability.
A few weeks ago an opportunity arose to reformat my laptop’s hard drive. I started fresh with a Fedora 25 live image installed from USB. Afterwards, I updated to Fedora 26 with updates.
I knew I was going to build Gautier RSS later on. Using the dnf utility, I installed various development packages and shared libraries and their headers. All has gone well so far but it was time to test.
Github is very useful. The source code for Gautier RSS just lives on there for easy download. I went to the repository hosted online and did a git clone to pull in the last commited version to the laptop.
The question is does the code build and run? I used the make file for the shared versions of the library dependencies. Two errors followed that revealed I had installed the wrong editions of Allegro and Gnome XML. I fixed that with dnf and tried the make build again. Success. The make build works.
What really piqued my interest in testing this project is the excellent book by Peter Gottschling. His book, Discovering Modern C++: An Intensive Course for Scientists, Engineers, and Programmers takes a different view on C++ that I find intriguing. I plan to later reconsider the Gautier RSS implementation based on the perspective he and others have presented. Namely, migrating from a std. 03 mindset to a post std. 14 mindset. The current and previous versions of Gautier RSS used std. 14 elements and suggestions from Stroustrup, Meyers, and others but was still slanted towards a std. 03 approach. Later versions will be based fully in a post std. 14 model.
Meanwhile, the results of the attempted build is shown below. The build errors are in red. Successful build comes after followed by a directory listing that shows the finished result. Last, I run the program from the build directory.
SpaceX to launch most powerful computer ever sent to space station
that is apparently an advanced HP computer. Looks like they are using HP laptops up there too. Good to know the R&D into more advanced computing is continuing and it is surprising how the supercomputer form factor is evolving. Maybe some of that advanced computer engineering will pay off one day with better laptops and desktops including operating systems. HP has regained momentum in non-handheld computer hardware.
You may have looked at a list of updates on a phone and wondered Are App Sizes Out of Control? Well, a few have weighed in on the matter and do see growing app sizes as ridiculous. App bloat is unnecessary.
A lean, clean simple app is pleasant. Launches quickly. Barely a delay between splash screen and ready to use. An experience so smooth you don’t notice it.
Many apps today are quite the opposite. Bloated and slow is not a permanent state of affairs. The way apps are built can be so much better.
Big capacity that beats external hard drives. Nice if it is affordable, available, and reliable over decades. An excellent way to keep information, at least that is the reputation of tape, over a long term. Time to reintroduce tape and in a form that is certifiably durable and dependable.