A Case Against FOSS

Questions, answers, and news related to computer hardware and software

A Case Against FOSS

Postby Yogi » 16 Jul 2014, 12:15

FOSS is the acronym for Free and Open Source Software. It's the reason Linux exists, but the idea behind FOSS extends well beyond community maintained computer operating systems. The idea is to make useful software available to the masses without cost to the end user. Transparency of the software source code lends itself to improvements and customization to suit the needs of any individual with the appropriate skills to make changes. This seems like a good idea in contrast to all the costs of ownership and maintenance of proprietary and copyrighted software.

What follows is a lengthy technically burdened saga about what should have been a simple task. But, as the old saying goes, “If you think the solution is simple, you don't fully understand the problem.” Herein I'll try to make clear why FOSS is not always the right answer.

My adventure started nearly twenty years ago when I connected my home computer to the Internet for the very first time. Earthlink was owned by the Sprint network in those days and had a reputation for being a reliable Internet switch. Part of the services provided by Earthlink was free space on their web server on which I could publish anything my heart desired. I created a couple web pages about me, complete with pictures and poetry and music. The music was a critical part of the message and it played automatically when you opened the web page. The audio player was provided by some now forgotten online purveyor of DIY code that can easily be embedded into a web site of your choice. Since I wasn't all that good at writing my own code, I went with what they supplied. It was free, simple, easy, and worked every time. Plus, it resided on their server.

Fast forward a couple decades (into a new century) where the audio on my personal web page stopped playing. It happened late 2012 or early 2013 and I figured it was just the player's hosting web site having some transient problems. After a few months of no audio I looked into it further and discovered the host site doesn't exist any more. That presented me with the problem of creating my own code to replace what was free and easy to copy from somebody else. I didn't bother to find out why they don't exist, but it soon became obvious. The software they offered was no longer free. A law suit involving audio players was won by the media player inventors who could now impose a licensing fee on anyone who tried to use their players. That is anti-FOSS, of course, which is why the site shut down. They were providing copies of software somebody else invented.

OK, it was just a matter of finding the right code and replacing what I had. How difficult could that be?

According to w3schools.com it could be VERY difficult to cover all situations. The first issue is that there are a number of ways to play audio on a web page. You can in theory:
-- Use Plug-ins via the use of HTML4 <object> and/or <embed> tags,
-- Use HTML5 with it's <audio> tag, or
-- Use hyper-links instead of audio players
Whatever the webmaster chooses, ideally it will play in all the popular browsers (Internet Explorer, Chrome, Firefox, Safari, Opera) and on all hardware (PC, Mac , iPad, iPhone). Unfortunately, there is no single audio player that will do all that.

The media player problems vary from simple to unsolvable. The file format in which the audio was recorded must be supported by the media player you want to use. You will end up converting your audio into at least three different formats to cover all situations. Then you must write some javascript code (you do know how to program in javascript, do you not?) to try and determine what browser is being used by the visitor so that you can server up the correct audio file. Then you must hope and pray that the web server on which you store your finished product supports the use of the files and codes you came up with. HTML5 was invented to get around these type of problems, but not all browsers support HTML5. Some that do put their own spin on it. So much for standards.

I use several browsers for various purposes, and all of them were employed to try and sort out my audio problem. Mozilla Nightly (a nightly build release of Firefox), Google Chrome, Opera, Maxthon, and Internet Explorer are all in my arsenal. I write HTML from scratch and use a text editor that is set up to send the output to my Chrome browser. At one time Opera was my web page default standard for development purposes, but they have gone down hill recently. So, now I figure if it works in Chrome, it will work anywhere. (bad assumption Yogi)

Knowing what I was up against, I composed a test page of HTML script to create media player calls from two forms of the <embed> tag (one specifying the media player, and one letting the browser/server figure it out), HTML5's <audio> tag, and an <iFrame> tag that certain browsers demand. Here is a screen shot of what it looks like in my Chrome browser (ignore the text at the bottom for the time being).

[ img ]

All versions of the media player work in Chrome. The EMBED CODE 1 is not functioning in the screen shot because I did not install the plug-in for x-mplayer2 into Chrome. (there are other mime types that could be used: x-ms-wmp, asx, x-ms-wma, x-ms-wax, and only god knows how many others)

The same HTML code running in each of my browsers rendered wildly different results. I was looking for one audio player that would work across the board and the HTML5 CODE did exactly that while I tinkered with it on my desktop computer. But, the real test would be from an online web server. I was shocked to see that Firefox no longer recognized the HTML5 code when this web page was sent from my Earthlink ISP's server. That server only recognized the iFrame code which did not work from my desktop by the way. UGHHHH! In desperation I fired up my Ubuntu Linux laptop and requested the page from Earthlink's server. Ubuntu Firefox recognized my request but would not play the music. They offered to download a copy of the music file for me to play off line in their FOSS media player called rhythmbox. Well, before I could play it in rhythmbox I had to install the browser plug-in which would not install because a dependency was missing. Fortunately, it knew what was missing and offered to install that too. After all that, rhythmbox opened and I could click on the audio file to hear the music, meanwhile forgetting why I came to this web page in the first place. I must note that both Ubuntu and Firefox are FOSS offerings that are playing by the rules. Neither software vendor recognizes any of the proprietary media players and will not let you easily use them in their software.

I then uploaded my audio test web page to the brainformation.com server. The HTML5 CODE started playing immediately in Firefox. All the other players started to work after the audio file was downloaded some 20-30 seconds later. Both Brainformation and Earthlink servers are running Apache web services, but brainformation has a bunch of modules added that apparently are not installed on the Earthlink server. Upon further investigation I found Brainformation servers are running UNIX server software with Apache, but I could not determine what OS is on my Earthlink server. It could be a Windows server for all I know.

At this point I had three different file formats for my audio: the original MP3, an MP4 version, and an OGG version. The choice of which HTML CODE to use was more complicated given that not all the popular browsers render HTML the same way. In order to cover the widest anticipated audience, I had to use both HTML4 and HTML5 in the final product. But then, it was incumbent upon me to determine beforehand what browser my visitors were using so that I can provide the appropriate code for them to hear my music. The text at the bottom of the screen shot shows how certain browsers identify themselves to the server from which they make requests. The userAgent is the salient information and you can clearly see that there is no standard for identification. I think I came to a reasonable solution but it all depends on the server side software and how it's configured. Obviously, there is no standard there either.

You may be asking, why am I so down on FOSS here. They are making the life of web page developers extremely difficult. Somewhere just under half the Internet is using Firefox browsers (FOSS). About the same number are using IE (proprietary). The proprietary software is working well enough to play my audio files across the board. Firefox does not and cannot by definition. I could use a hyperlink in Firefox so that my visitors can click on it and hear the music, but who would go through that trouble? The FOSS people are not for profit and it's against their principles to actually pay for anybody's licensed software regardless of how poorly their own performs. It's illegal for them to do anything else. They are shooting themselves in the foot, and it seems particularly true when Firefox is used in a FOSS operating system such as Linux. It's a nightmare that will be resolved by web page designers as time goes on. They may stop developing for FOSS compatibility. HTML5 holds a lot of hope going forward, but as of today FOSS proponents haven't figured that out yet. Hopefully they will get the message before Google takes over the world.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 17 Jul 2014, 12:15

Wow Yogi, you sure have the problems don't you!

I guess if you are against FOSS, then you are also against Net Neutrality.
And want everyone to PAY for EVERYTHING.

I had a music file on my first original website also.
It was not set to any specific audio player.
I let the viewer of my website use the audio player of the THEIR CHOICE, whichever they had installed.

From your rant, it appears you are trying to Embed the Player of your choice?
Forcing the viewer of your website into what you want them to use?

One thing you failed to mention was that for the past two decades, when we wrote for websites, we ALWAYS had to write one page for ALL BROWSERS and one for PROPRIETARY Internet Explorer, because Mickey$oft had their own UNSHARED PROPRIETARY Codes.
They only came around in recent years when the finally realized, web programmers were no longer catering to their nonsense.

Now Mickey$oft is doing it again. They are TWISTING XML code around and adding Mickey$oft PROPRIETARY CODE into XML so that only their Proprietary software can read their Proprietary changes to the XML code base.
They are even changing the coding for UNICODE and ASCII into their exclusive Proprietary Formats.

Although you use FOSS, I wonder why?
When it appears you want only one MONOPOLY to control everything.

I do agree there are many things not compatible in the Open Source world.
That's fine with me, it was never meant to be compatible.
Oh by the way, neither are things in your Proprietary World.
My Apple II hard drive disks will not work on a MAC or a Windows PC.
Can you imagine that. Apple and Mickey$oft are not compatible with each other.
And you have to write separate CODE for each of them. Amazing, I wonder WHY?

OK, so I'm being a bit sarcastic.

Personally, I would like to see the ability to patent or copyright all software code completely abolished.
WHY? Because every last bit of software coding is written using FOSS Source Code.
If you write your new software program, using C++ for example. If you are going to CHARGE FOR, Patent, or Copyright your code, THEN for those same reasons, C++ should be Patented, Copyrighted, and SOLD to you to let you use it. With ongoing royalties for writing your code using THEIR Developers C++ Source Code.

Over 99% of Proprietary Software is written using FOSS at their very base.
And guess what, their Proprietary Software is NOT COMPATIBLE with anything except their OWN equipment, in most cases.

How much did you pay to obtain or use the HTML code to write your website? NOTHING you SAY!
Why not? BECAUSE IT IS FOSS! But you are against FOSS.

Unless you are writing all of your code in direct BINARY, then rather than bashing FOSS, you should be thanking FOSS for all the Programming Languages in use today.

Think about what life would be like, if ALL Programming Code and the Software generated from its use, was Proprietary and cost an arm and a leg each step of the way.
You would have to pay for the use of the OP Code, the Assembly Code, the Compiler Code, etc. etc. etc.

I rest my case...

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Kellemora » 17 Jul 2014, 12:20

Just thought of something.
On one of my websites, I decided to auto-load PDF files for viewing when a visitor comes to my website.
The same as I did with my music, I only embed the PDF file, NOT what Viewer they should use.
I let the user use whatever Viewer they have installed on their computer, whether it be one from Adobe, or a 3rd Party Viewer capable of reading and displaying PDF files.
This save a LOT of headaches, as I do not have to first determine what browser they are using and what PDF Viewer they have installed. And if IE does not recognize W3C standard xhtml/css STRICT, that's the IE users problem, not mine. I no longer kowtow to Mickey$oft or their users.
My files are W3C Error Free. I can't waste time trying to work around those who have machines that are not compatible with the recognized standards...

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 19 Jul 2014, 08:28

Most people I talk to about FOSS have no idea what it is or why it might be a good thing. They just go with the software that suits their needs and will use it freely, pay for it, or pirate it, whichever is easiest for them to do. In the case of Firefox the FOSS approach has proven it's value and stripped Microsoft of it's monopoly over Internet browsers. The problem I see consistently is that the goal of FOSS developers seems to be to merely show that they are better because they are free and not constrained by silly things such as copyrights and patents. Creating a superior and easy to use product is given lip service but seldom implemented. My story about how FOSS fails in my application is not unique. The ranking of Linux in the noise level of desktop operating systems for decades further demonstrates how the FOSS approach is not delivering what they promise.

It can be argued that there appears to be chaos among the paid-for software vendors as well. That problem is being addressed in HTML5. In my recent experience every vendor of Internet browsers executes the new HTML flawlessly, except for Firefox which is FOSS based. Why is Firefox deliberately choosing not to comply with the new W3C standards? The answer is that they cannot comply without infringing on established copyrights. It remains to be seen if they can catch up to the rest of the crowd but Mozilla obviously thinks they can. They have now stepped into the operating system realm with Firefox OS.

There also are other issues I have with FOSS. Open source code exposes software vulnerabilities to your competitors, not to mention every hacker on earth knows what is inside your creation. The truth is that in this country there is no better motive than the profit motive. Protecting profits is what copyrights are all about. It's true that the consumer pays those profits, but they also benefit from the incentive developers have to do better than the rest. The better their creation performs, the more profit they derive. FOSS removes that powerful incentive to create superior products. Each software developer must make a choice: the ideal to be free and open, or to have money in the bank from their profits. Which would you prefer?
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 19 Jul 2014, 11:28

I agree with your points Yogi...

But there are other aspects of FOSS you haven't mentioned, such as the License Agreement to how FOSS may be used. It is these License Agreements that was supposed to prevent some of the problems you see today, with so many versions of the same thing.

You brought up vulnerabilities, so I will expound on that aspect, and point out why FOSS is actually better than proprietary.

In a proprietary atmosphere, only a select few programmers work on the coding for a program. No User knows what is in that code, and they have no Eyes watching their back for them either. You have to trust the provider of your software that they closed up the loopholes.

We all know how well THAT method works, by the amount of hackers and anti-virus software being sold for cleaning up the damages to proprietary software products.

Open Source software has thousands of Eyes watching your back. It is almost impossible to slip malicious code into recognized software that all the programming guru's are keeping watch over.

By the same token, since it is open source, a hacker could publish malicious code and put it out there under the guise of the original, or a deviation from the original, and some users will download the free program.
I've also seen this happen many times with proprietary programs as well. Some company will publish a malicious code under the same filename.exe as a known popular program, and get to many.

To help prevent this in the Linux world, we have repositories, we know we can trust, as well as those we know to be leery of. We are free to point our package manager at one of these leery repositories, at our own risk.

I've downloaded programs off the web, because they were not in our trusted repositories and I needed that particular program. This is where you have to be super careful on where you download from. You may think you are going straight to the programmers own website, when some hacker made a clone of his website.

Plus you brought up Firefox not being able to bring their web browser into W3C compliance, due to copyrighted programs needed to make it function with new additions to the code other browsers are using.

To the best of my knowledge, W3C will not standardize any feature that is not available for insertion into all browsers. In other words, something proprietary to a single company cannot become an industry standard.
It can become a specific user base standard based on consumer usage, but cannot become standardized as far as the computing world is concerned.
A good example of this is msWord. Due to the large base of Windows users, msWord has become the word processor of Windows users choice. Simply because it is proprietary, it is not the standard word processor for Apple, Mac, BSD, RedHat, Android, or GNU users. In other words, it is not universal across all platforms.

As an aside: I was reading an article the other day about Androids failure to comply with both of the GPL licensing requirements and Linus Torvolds Linuz agreements for using the maintained Linux kernel.

Software written for use on computers, regardless of the OS being used, can be proprietary, closed source, or open source. But one cannot take open source and use it for resale by adding closed source to already open source code, or making it proprietary.
This has created quite a dilemma in the open source world, because if a code is proprietary, no one can see what is in it, to see if they are using open source code with modifications.
Several proprietary apps sold for Android were discovered built from open source code and being sold as proprietary, totally ignoring the licensing allowing them to use the code.

Because of this, I think we are going to see a big shift in the tightening of open source and the dissemination of proprietary and close source codes to either industry appointed or federal examination committees.
We both know this will be a major nightmare to the computing industry, because so many codes are so close to the same. Like car tires. Every car tires tread design is patented, and the difference between many of them is no minute that are nearly identical. I see computer codes becoming so cut n dried that changing one symbol renders the entire code as a different code, so it can become proprietary.

It is this type of restrictive nonsense that will kill the computing industry, unless they completely abolish the current copyright regulations the way they are written.
Think of the number of books and novels that have been written within the last 75 years. Unless a good portion of a new work is identical to an existing work, it is not considered plagiarized, or in violation of the copyright.

These same rules for copyright as applied to books cannot be applied to computer coding, because of proprietary and closed source coding. So how do you know if one programmer is plagiarizing another programmers code, when you are not allowed to see it?
For this reason, the same as with Patents, if it is not disclosed so that someone familiar with the art can duplicate it from the Patent when the patent runs out, then it cannot be patented.
Many of us feel the same should apply to copyright. How can you copyright something not disclosed to the copyright office to see if it is even copyright-able?

If you were a programmer, and suspected a proprietary company stole your code.
How could you prove it, if you cannot see their code?
How would you know if YOU were violating someone else's code, if they keep it hidden?

I hope you see the dilemma and problems proprietary and closed source coding is causing the world.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 19 Jul 2014, 13:45

[ img ]

    #thinker {
    position: relative;
    top: 225px;
    left: 103px;
    }

Not to belabor a point already made, but ...

I've been working on a redesign for our home page. Each element on the page is an object onto itself so that I can specify exactly where I want that element to display. The image above shows how "the thinker" graphic is rendered in the current version of Firefox (version 33) compared to how it is rendered in the latest version of Internet Explorer (version 11). Between this paragraph and the image are the instructions telling the browsers where to put the image. Relative to the container it is in, I want the upper left corner of the image to be placed 225 pixels from the top and 103 pixels from the left side. How to do this is not rocket science and is clearly spelled out in the W3C standards for web browsers which Mozilla agreed to along with Microsoft and a lot of other folks. There is nothing proprietary about the instruction, plus I ran my code through a certification machine and it is totally compliant with W3C standards. Yet, Firefox stands alone among five browsers I used to check out the end product. As an aside, I also viewed this in a stripped down version of Firefox called Iceweasel running on my installation of Linux Debian 6.0.x. The thinker graphic is even further misplaced to cover up the text in the footer beneath it. I thought the two negatives would cancel out, but it only increased the error.

I am willing to accept that the FOSS philosophy might not be the culprit here. But, Mozilla and the Linux operating systems hosting their browsers do not render my web page as intended. All the other propitiatory browsers do render it correctly in MS Windows. Obviously, in this instance the Mozilla organization as a whole is contributing to the chaos by not implementing the W3C standard that it agreed to follow.

The montra saying that open source is more secure than proprietary software is an exercise in deception. Hackers don't bother to write malware or virus code for open source software. They don't have to. They exploit vulnerabilities in the code. In the eyes of FOSS vulnerabilities are somehow not as much as a security threat as are viruses or malware. Just read the newspapers about how many corporate databases have been hacked and billions of records of personal financial information put at risk. That happened due to system vulnerabilities that are the envy of every virus hacker out there. Additionally, what do you think is easier to do? Write a virus for software that is viewable by anybody because it is open sourced, or to write a virus for a black box of code that nobody knows what is inside?

The arguments we are rehashing here are a old as software itself. Personally I like the idea of free software that I can modify as I see fit. I simply think that the FOSS community has lost it's focus on it's mission to create products that are valuable and easy to use. They are instead trying to defend themselves from the higher priced system, and frankly FOSS seems to be losing the battle.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 20 Jul 2014, 10:16

Hi Yogi

It is known RELATIVE, when used in a container, is not recognized and/or rendered differently in different browsers.

Why are you not using ABSOLUTE? It's all I will use!

.header {
background-image: url("../gfx/header1.jpg");
position:absolute;
top: 8px;
left: 175px;
width: 80%;
color: #660000;
background-repeat: repeat-x;
border: 5px;
border-style: ridge;
border-color: #660000;
padding-top: 15px;
padding-bottom: 15px;
text-align: center;
}

I've found a few glitches also, where things do not work as they should, and had to find workarounds for them.
One example is the float image on each page. I cannot control the page length using a percentage as it is supposed to work, so I moved the height to the html page and set each page manually using inch measurements.

#floatimage {
background: url("../gfx/dutchrhudysilhouette.png");
background-repeat: no-repeat;
background-attachment: fixed;
background-position: 0px 10px;
width: 100px;
} /* height is located on each individual html page to fix the 100% bug */

Here is the html line.

<div id="floatimage" style="height: 25in"> <!-- moved from css stylesheet to here to fix bug -->
</div> <!-- closes id floatimage --> <!-- because auto nor 100% work properly -->

And the above is placed BEFORE I open the CONTAINER, otherwise I have other problems on different browsers.

I don't use XHTML/CSS enough to remember much, and have to go back and study again each time I want to do something or make changes.
Started looking into HTML5 so I'm ready when the time comes. But it appears they still have tons of bugs to work out of that version yet before it's really usable.

Good luck with your image.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 20 Jul 2014, 14:43

My point in this thread is that there are standards to which all the major browser developers have agreed, here, in this century, ca 2014. It wasn't always that way. I can't speak for Safari, but all the other popular browsers and a couple obscure ones render my compliant CSS styles as intended with the exception of Firefox. Firefox is the only one in the group of browsers I have that is FOSS.

Absolute positions is a fantastic brute force format that has it's place when positioning containers, for example. Relative positions work as expected inside the containers. You may have to reset your web page for that to happen:
When a fluid web page is being rendered absolute formatting is useless because the absolute positions do not scale up or down along with the changes in browser window size. Relative positions are the only way to go there. My original thoughts for our home page were to make it fluid.

Another consideration that must be addressed these days that wasn't an issue when the first home page was developed is the use of mobile devices. Some mobile screens have a maximum resolution of only 760px in width. Again, relative positioning would seem to be better suited when trying to offer a single page that will be rendered as intended on both devices.

Well, I've given up on trying to make a single page for both worlds and have resorted to a fixed width webpage design. Relative positioning still makes sense within the major containers but placing the containers themselves is an "absolute" necessity. I have two landing pages now with the default being the one intended for desktop/laptop viewing. The mobile page is uploaded to the server (http://brainformation.com/index_mob.html) but I have yet to figure out how to redirect people using mobile devices.

You have my sympathies regarding length of a web page. I learned long ago that there is no such thing in practice as a 100% length dimension. The containers on our home page have a defined height that works very well for my purposes. I have no trouble counting pixels when I have to, and besides, the current page (as of this morning) is not intended to be of a fluid design.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 21 Jul 2014, 10:17

Hi Yogi

I may be wrong about this, because it has been a few years since I set up my webpages.
But using absolute only sets the top left corner in the manner in which I use it.
A width instruction as a percentage adjusts it to the pixel width of the display.
When I first designed my web pages, I checked them through
browsershots.org
and they displayed as expected on all of them, except perhaps one or two I never heard of.

On some of my pages, I forced a maximum width, because of wide-screen monitors distorting images.

Each of my computers here are set to a different resolutions, this one I just changed from 1024/768 up one size, but still square, even though this is a wide-screen monitor. Mainly because I do graphics work on this computer and I need circles to be round, not oval looking.

I've seen my websites on everything from laptops, netbooks, tablets, and even on the Kindle Fire, which is technically a tablet, and it all looks great. No text overlaps images, or lands on top of itself, as I see happen on so many websites these days. Almost every website where I see problems, and check the source code, they are using Transitional which allows many errors to go unnoticed, or the use depreciated coding, some of which is no longer recognized by current day browsers.

I have to laugh at one of our local writers' guilds. They do their page in msWord, then save as html and post it on-line. Needless to say, it has tons of problems.

There are many things I could never figure out how to do, and/or it would take php or js to accomplish.
Like my Paged Index box on the left. It would be nice if I only had to have this once, and make it appear on each index page. But I never could figure out how to do this and get it to work right on all the major browsers. So, because I couldn't figure it out, I ended up placing the code on each indexed page. This of course means if I make a change to the landing page, I must go to each indexed page and add the changes there too. This makes for a lot of extra work each time I want to change something in the Paged Index box.

I'm sure you will get your new code worked out and running smoothly. It took me forever, not being experienced, to get mine where I was happy with it. Then since I don't do it every day, I forgot 90% of what I learned while building it.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 22 Jul 2014, 08:36

For purposes of demonstration, let's assume I'm placing an object in the center of your 1024x 768 monitor. Assuming we are using full screen to view it, the absolute position would be 512px from the left and 384px from the top. Now let's look at it on my 1600 x 1200 screen. Obviously the object will no longer be centered. The problem becomes more complex when a page with multiple columns is being displayed. You don't want the content of the right hand column to be in absolute positions. Positions should be relative to the right side column container origin instead.

There are many tricks of the trade, but some browsers make those tricks tricky to implement. Using a strict document type is a good idea in some instances, but we both know that not every browser complies with the strict W3C standards. Even when they comply the way they render standard features could be different. For example, the 0-0 upper left corner is not always counted from the edge of the browser window. Some browser start counting a pixel or two inside the window to account for the browser's intrinsic border. Getting to the x-y coordinate is standard and all browsers comply with that, but not all browsers agree on where 0-0 is located.

I'm not very experienced with designing web pages and can't think of a good way to make your paged index work without using something like php and an SQL database. This website is doing what you aspire to accomplish but in a slightly different format. The pages are all templates here, but the content varies. The content is extracted from our database and the pages are build dynamically as they are called by the user. Any changes to your index would only be made only once if you were using a similar database scheme. Take a cue from all those nuisance ads you like to block from web sites. You can do that because they are being displayed from a remote server whose content changes continually. They are not hard coded into the page. In essence the ads are in a database and could be placed on any number of different web pages with the proper call to the database.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 22 Jul 2014, 10:24

Hi Yogi

In the case of placing something in the center. I don't think I would use absolute positioning, for the very reasons you mentioned.
Although I don't remember how I did some things. If I recall, you can have more than one container, one inside the other, and place one in the center using the code center if it exists.

One thing I do remember from when I was doing my own web page, is I had to purposely leave certain things outside the container for them to work right.

Way back when I was first starting to build the Indexed Pages List on my website. I had a single separate html page that contained the data, and called it on all the index page as well as other pages from this single file.
It did work, but some browsers messed it up royally. What made it work on one browser, caused it to display differently on other browsers. I never could find the right combination of codes to get it to work right, which is why I went to adding the code to each indexed page.

I'm probably using the wrong terminology here. I have the main Landing Page we all have "Index.html"
Everything else stems from this Landing Page via links to my first tier of html pages, I call these "Indexed Pages," because you get to them from the Main Page.
In my Paged Index box, are the Topics that take you to these individual pages.
Once you are on an Indexed Topic Page, you may find links to various sub-topics.
These I keep in a separate folder named "HTML Pages." They DO NOT have the Paged Index box on them.
Naturally I have images stored in a GFX folder and the CSS pages in a STYLE folder.

On my Main Landing Page, if you scroll down, you will find the Paged Index repeated, complete with Topical Links as well as Links to the sub-topic html pages. This whole extra section to the same things, was added as an enhancement suggested by the IT manager where my frau worked before she retired. All of their company websites have pretty much the same concept on them. Except they have buttons across the top, where I have the Index down the side. At the time, they did not have drop down boxes under the top buttons. They do now so changed their pages considerably since I first set mine up.

Regardless of what browser I'm using, I run across several problems with drop down boxes on many of the websites I visit. On some, the drop down boxes are behind the other things on the page near the top. On some, when you roll over the button, the whole button bar drops down to the page and you can't click on anything. Or the text that is supposed to be in the drop down box, you only see the last word or something else just as strange. So I figured it was best to just leave what I have alone, since it works.

One thing ODD I did notice the last time I went to make some changes, and went to the various teaching websites, it seems all of them have dropped XHTML, including W3C's teaching pages. They only have HTML or HTML5 training pages now. My old saved link to the XHTML pages, now bring up HTML5 pages.
Makes me think after going to the trouble of doing my site in XHTML vs HTML, that I will have to change them to HTML5 sooner than I thought. Which means learning all over again.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 22 Jul 2014, 12:32

I'm guessing that you are having all these navigation problems with drop down menus on your Linux boxes using Firefox. If that is true, it only lends support to the subject of this post. They don't do things like everyone else over at Mozilla. Try using Opera to see if it helps.

You also bring up a couple other concepts wherein standardization seems to be a problem for certain browsers. Containers are typically defined by the <div> tag in HTML. When I first started working with HTML I thought containers made sense. You put little boxes inside big boxes and then arrange them to look pretty. It's not that simple in reality. Containers are only a concept. The containers inside them can exceed the dimensions of the parent container. Thus, the parts can become greater than the whole. When you try to "float" objects, use margins/padding, or positions (relative and absolute), objects that conceptually should remain inside the mother container do not do so. Position instructions, for example, are not inherited. And, of course each browser has it's own way of screwing up these overlapping <div> denoted containers.

Browsers can be three dimensional. The z-index CSS instruction is used to control what overlapping sequence is used in a given situation. Take my thinker graphic from above as an example. The image overlaps the footer. One way to solve this problem is to assure that the z-axis placement of the image is in the background, or at least has a lower priority than the surrounding objects. Thus, I specified that the thinker have a z-index: -1; meaning always on the lowest layer. That put the footer on top of the thinker no matter where the graphic ended up on the page. It was a good solution in all browsers except Firefox. They don't recognize the z-index instruction yet it's part of the W3C standard.

HTML5 is supposed to solve all these problems and more, my experience with embedded audio notwithstanding. Not everyone is up to speed with HTML5 yet so that there is still hope for Firefox complying with the rest of the world when they finally reach the 21st century. I've not used HTML5 until a few weeks ago. W3C schools does a good job of explaining things and I'm not sure I'll even have to buy a book to come up to speed. If you know enough HTML4 the next generation is just a more logical implementation. Of course there will still be room for misinterpretation of the standards, but hopefully everyone will at least be pointed in the same direction.

One trick of SEO specialists is to include a link to a "site index" on the landing page (index.html). The crawlers will go to that site index page and index every link you have there, which of course would be every page on your web site. The other advantage of a site index is that it can be used for navigation, although it's not a good practice. Too many clicks for the drive by viewer to bother with. From a webmaster point of view, the site index is a single page to maintain instead of several pages with the same information. I'm pretty sure you can use an iFrame tag to place that site index on every web page on your site, but I'm also pretty sure iFrame is being deprecated. Then again, this would be a good argument for using a transitional DOCTYPE instead of strict. The deprecation process is many years in the making. You will have a perfectly viewable site until the old instruction fades into the background by attrition.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 23 Jul 2014, 10:55

Hi Yogi - Although Firefox used to be my favorite eons ago. I've been using both Chromium and Google Chrome for a long time now. Google Chrome is the browser I open to do things, but e-mail links open Firefox. I have it set that way on purpose, as I have no log-in data stored in Firefox.

Ironically, it took me years to move away from Cello, if you remember that one. It did not handle images so all I got at my end was the text portion of web sites. This was perfect at the time, working on dial-up, speed was important to find what I needed as fast as possible.

Although I've used Opera and have no complaints with it. Some of my add-ins, especially a toolbar I use at work, does not work in Opera. They want us to use Firefox at work, but the Linux version of Firefox is not the same as the Windows version. So I use Google Chrome and can do what is needed.

I played with z-axis trying to get a row of images to overlay and line up like a spread deck of cards. It worked on one browser, but the individual images were not clickable. On another browser, the images were clickable, but not spread out the way I wanted and shifted left. I probably played with this for over two weeks before figuring out all browsers don't recognize the code.

I'm hitting one heck of a lot of web pages lately that have NO DOCTYPE at the top. I'm thinking either the html pages are generated from php, js or something for display, or perhaps HTML5 requires no DOCTYPE. I've not studied it further to find out why.

On personal web pages, I find a lot that are done using something like msWord, then saved as html and uploaded to their websites. Lots of lines of useless code in them that do nothing except take up space.

Although I don't play games anymore, too busy. I almost had to play them on Firefox in order to get the Gift Links in bulk, rather than opening a separate page for each one, like in Chrome or Opera. So their History File is far advanced from all the other browsers in ease of use and the ability to organize and copy all links at once.
This is one of the reasons I have Firefox set as the default from my e-mail and from clickable links on Farcebook.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 23 Jul 2014, 14:47

HTML5 has only one DOCTYPE: <!doctype html>

I've seen warnings on Microsoft tech support sites to the effect all HTML read by Win8, Win7, Vista, and XP must have some DOCTYPE declaration. MS can't guarantee IE will render the page correctly without one. I suppose you can make a web page without a DOCTYPE declaration, but why would you want to? PHP will generate what you tell it to. Leaving out the DOCTYPE declaration is a major faux pas in my opinion. It's bad enough out there when you write compliant code, but not telling your browser what kind of document it's about to handle is insane.

I've looked at the userAgent string for Chrome and Chrome-like browsers and it is fully loaded with information about what the browser can handle. Look at my graphic further up this thread and you will see Firefox is naked compared to the rest of the browsers. Admittedly the userAgent string is informational only, but it does say what the developers want you to expect from their product.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 24 Jul 2014, 12:09

Hi Yogi
That is what I meant.
Rather than seeing the doctype spelled out, like this common one
Code: Select all
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">

or one for strict like I use
Code: Select all
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">

or like this generated one from Google Chrome
Code: Select all
<!DOCTYPE html><html manifest="/_/chrome/newtab/manifest?espv=2&amp;ie=UTF-8" lang="en">

What I'm finding on a few pages is only
Code: Select all
<!DOCTYPE HTML>

with no other information behind it to define the type.

Seems to me with no other information, the browsers wouldn't know what to do with what they are presented.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 24 Jul 2014, 13:47

That last DOCTYPE is the format for HTML5 only. When a browser sees that, it applies all the HTML5 rules. Anything deprecated in earlier code may not render correctly or not render at all. I'm not sure if the HTML5 innovations are available if you use any of the other DOCTYPE's, but I'm guessing they are not given they were all intended for HTML4.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 25 Jul 2014, 08:25

Hi Yogi

Supposedly, a website done in XHTML, especially in STRICT, is fully compatible with HTML5 as it stands.
HTML5 allows for more sloppy coding practices, and adds many new features not found in HTML4.

HTML5 eliminates the need for plug-in's, and being case insensitive one is not restricted to all lower case, as in XHTML. But at the same time, they claim XHTML maintenance is easier than HTML5. Why I don't know?

My original website from eons ago used HTML$ and had frames and tables, which are no longer used. Although even back then, we were not supposed to use tables for formatting purposes, in many cases, it was the only way to get a page to display the way you wanted it to.

Before I started studying to rewrite my websites, I asked at least twenty major website maintainers, and nearly all of them said to go with XHTML, because it was what newer html versions would be built from. Even then, HTML4 and HTML4.01 had several tags slated for depreciation, to force use of CSS.
For me, this was like a nightmare at first, not only did I have to learn html all over again, now it was XHTML/CSS, two things to learn that work together.
The only person who had time to look at my hand coded pages, which I used Transitional at the time, told me to move up to Strict, since I had nothing in my pages to dumb it down in order for it to work. I only had to make like three or four minor changes for it to pass the W3C strict test, and if I recall, these were just closing tags.

If I can find the time, out if curiosity, I may change my DOCTYPE and run it through W3C as HTML5, just to see what happens. That way I will know if the things I've heard or read are accurate or not.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 25 Jul 2014, 08:56

The drawback of HTML5 is clearly demonstrated in Firefox. I don't know for certain if it has anything to do with the FOSS philosophy, but Mozilla's treatment of audio smells like that is the source of their incompatibilities. And that is just Firefox. I'm sure there are other browsers which are not up to speed with the latest coding standards. Unfortunately, even if all the browsers were 100% compliant with today's HTML5, there would still be major issues rendering web pages correctly across all platforms. All those Windows XP people don't have modern browsers, for example, and as you pointed out earlier Firefox for Linux is not the same as Firefox for Windows. There there are the mobile devices which live in a world of their own.

I never bothered with XML. My experiences in programming have engrained certain good practices that are inherent in the XML coding standards. And that brings up yet another opportunity for error; not every webmaster codes things the same way regardless of the standard they are using.
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Re: A Case Against FOSS

Postby Kellemora » 26 Jul 2014, 12:22

True and True again...

Printing and publishing companies that handle digital files to produce their printed pages are facing several nightmares due to Mickey$oft adding their own proprietary commands inside of otherwise standard XML formats.

A few smaller companies who handle POD printing services and e-books, have found a few of the new hidden non-standard codes, and wrote sub-routines in their converter programs to handle them. But like everything else, since the offending program DOCX is proprietary, the solution is only a series of guesses until they pinpoint each problems DOCX causes.

The larger publishers, like the one I work for, will not normally accept a DOCX document. They will, but clearly state they will convert the DOCX to DOC and return it for a proofread before consideration.

I'm sure you've seen on places like Farcebook or Twitter, posts where the apostrophe is replaced by a black diamond with a question inside. DOCX does not use the standard unicode for an apostrophe. So when you cut n paste to a post, nothing yet deciphers the wrong code.
I don't see why a program should have to look to see if the XML provided was butchered by Mickey$oft and just let those who use DOCX see what a mess it really causes.

What was Mickey$ofts purpose in changing a code that has been standard now for over forty years?

Nobody Knows!

Unless they are trying to change UNICODE enough to make it their own proprietary format.

I don't think they have enough users left in high places to force the entire industry to kowtow to them anymore.
They haven't swayed the Big Five Publishers to change. Nor have they technically caused a change at Smashwords. Smashwords still uses their same meatgrinder. A DOCX is converted to DOC before being sent through the meatgrinder. Same way at Draft2Digital. CreateSpace only accepts PDF, DOC, or standard XML. Some will accept clean versions of HTML, whatever that means, but with heavy limitations. Most submitted HTML version are kicked right back as not clean. I don't know anyone who managed to get one through yet.

As an aside: Iceweasel is supposed to be the Linux version of Firefox. Although I rarely use Iceweasel for anything, I have a few programs that open Iceweasel instead of Firefox. It appears to use the same add-ins and plug-ins. But I found several things that worked on Iceweasel that did not work on Firefox. Namely some news station videos. If I link to the video using Google Chrome, it works. If I link using Firefox, all I get is a black box. If I link using Iceweasel it works, and works better than Google Chrome. If I link using Chromium, I get the same black box I do with Firefox. This makes no sense, since Google Chrome is built on top of Chromium, they just add a few extra bells and whistles, at least in the Linux version. Google Chrome on my wife's XP box, does not have a few of the features I have on my Linux box, but I don't know if that is because of add-ins, or if she has those features turned off. I rarely touch her computer.

TTUL
Gary
User avatar
Kellemora
Brainiac Class Poster
Brainiac Class Poster
 
Posts: 3389
Joined: 07 Jul 2012, 18:52
Location: From St. Louis, current Knoxville, TN

Re: A Case Against FOSS

Postby Yogi » 26 Jul 2014, 13:46

It's a rare occasion when I have to deal with a .docx file, but I'm fairly certain I can do it from LibreOffice and it was even built into OpenOffice. Regardless, you don't need anybody's word processor to read .docx files. There are stand alone readers from Microsoft:
http://www.microsoft.com/en-us/download ... .aspx?id=4
http://download.cnet.com/DocX-Viewer/30 ... 79715.html
The only downside is that the readers need a Microsoft OS to work.

I have a friend in the graphic arts business and for the last twenty years I've heard stories about how digital graphics is a nightmare. What you see in reality looks different on your computer monitor, which is different than what your printer puts out. It appears that there are no standards in the publishing field when it comes to digital storage and reproduction. It all has to do with the Not-Invented-Here Syndrome and proprietary methods that are necessary to avoid copyright and patent claims. The only reliable solution appears to be to revert back to manual paper copies of any material for publication.

The FOSS philosophy would seem to solve the problem, but what publicly owned company wants to operate for no profit? If nothing is to be gained because all software is common, then I also see the elimination of an incentive to do better. Why bother if you get nothing out of it?
User avatar
Yogi
Oracle Class Poster
Oracle Class Poster
 
Posts: 7013
Joined: 04 Aug 2007, 19:37
Location: Chicagoland

Next

Return to Personal Computers

Who is online

Users browsing this forum: No registered users and 2 guests

cron