Free and Open Source Software(FOSS) for Jobs

[mamongato]

I believe government can create more jobs by utilizing FOSS instead of off the shelf ready software. Graduates will be employed to improve the existing FOSS and learn hard core development skills at the same time. Commercial software are expensive and they do not give money to our country but to other countries, we get only VAT from their sales. Their cost can be used to pay people who will build from scratch or improve existing FOSS. These software will be secure because loopholes will be closed as they happen, we will not wait for late patches.

 

[SauRoNZA]

How are hardcore development skills more prevalent whether commercial or FOSS?

A developer is a developer regardless. You won't have more developers because of FOSS.


FOSS software is often cheap in initial costs but expensive in time. Its not black and white that one is necessarily cheaper overall than the other.

You make it sound like we don't produce any commercial software here, which simply isn't the case. Sure we don't have our own commercial operating systems, but that's really a small piece of the pie.

And no we don't just make VAT. The government makes VAT while every link in the chain makes a profit of distribution and support. With FOSS they make nothing off the software and just stand to make anything from supporting it. Seems like less money and less jobs to me.

How will they be more secure? Closing loopholes as they happens is hardly security, that means it's wide open until someone finds the loophole, it could take years. In this case the larger the username the faster problems are found and patches....therefore chances are better that larger and possibly commercial options are more secure by default.

I'm not saying let's not do FOSS. By all means, let's do it. But I think your views and motivations are way off and rather subjective.

 

[mamongato]

Just to extend my post, it will also save government a lot of money. 1 office suite license may be 1500 X 5000+- govenrment employees, a lot of money spend right there instead of R0 LibreOffice for insttance you can just download and install. It will e more secure because you can see the code yourself so you can fix it. Now we just rely on patches which we are not even sure if they only download fixes or do they also still and upload our files to that manufacturer mainframe for data mining.

 

[Other Pineapple Smurf]

You are forgetting training costs. Also they buy the licences at massively reduced prices from the suppliers. Don't think government and big corporates pay the retail costs you see in the stores.

You are also making terrible assumptions about FOSS in the real world. The reality is that often open source cannot meet one's needs and paying for a closed source product is many times the better option.

You use LibreOffice as an example and while it is my only Word Processor / Spreadsheet I use, it still lacks an effective Powerpoint application (its implementation is awful). And there is still no substitute for Outlook.

 

[mamongato]

You say is discounted but still they are paying isn't it. LibreOffice has Impress and does all the things you will need in a presentation. You said Outlook, lol there is thousand email clients and I use them without any problems. There is Kmail, Evolution and many others. I know one institution that went completely OpenSource. The National Library of South Africa, with over 550 PCs and servers they are reaping the benefits. You have to make time to go there and see for yourself.

 

[DA-LION-619]

You say is discounted but still they are paying isn't it. LibreOffice has Impress and does all the things you will need in a presentation. You said Outlook, lol there is thousand email clients and I use them without any problems. There is Kmail, Evolution and many others. I know one institution that went completely OpenSource. The National Library of South Africa, with over 550 PCs and servers they are reaping the benefits. You have to make time to go there and see for yourself.

And cost for training and support?

 

[xumwun]

And cost for training and support?

I'm sure that's applicable to commercial software as well.

 

[DA-LION-619]

I'm sure that's applicable to commercial software as well.

Not true, switching software means the company will have to train staff, also there's more places offering Office support not that much people need it.

 

[Beachless]

Umm government has a big FOSS policy:
http://www.sita.co.za/FOSS/FOSS1.html

Its struggling like everything else though.

 

[mamongato]

And cost for training and support?

Training who, users want to TYPE and PRINT and LibreOffice and other FOSS oofice suites do just that. You will be amazed users adopt without issues. Anyway a company has to refresh its staff time and again on how to use the computers applications irrespective of a software. So the refreshment workshops will still continue as usual.

Companies spend a lot of money on a complete office suite with all the apps but users will be using only a word processor 90% of the time.

 

[garyc]

Umm government has a big FOSS policy:
http://www.sita.co.za/FOSS/FOSS1.html

Its struggling like everything else though.

This initiative was launched with a fanfare and then faded away. It looks like one of the main problems was the planning assumed that they would launch it on the desktops first and then look at the back-end enterprise stuff later. Saw some of the results of the initial roll-out, PCs that were unable to connect to network drives, email, calendar, etc. Needless to say this initiative died quickly.

It was the sort of SNAFU caused by it being planned by people who had no technical insight. Similar efforts in other parts of the world have been more successful.

 

[mamongato]

The failures with the SITA may have been caused by 5 year performace based contract thing, or perhaps realised there wont be enough "kick-back" money from suppliers once they go FOSS.

 

[Beachless]

This initiative was launched with a fanfare and then faded away. It looks like one of the main problems was the planning assumed that they would launch it on the desktops first and then look at the back-end enterprise stuff later. Saw some of the results of the initial roll-out, PCs that were unable to connect to network drives, email, calendar, etc. Needless to say this initiative died quickly.

It was the sort of SNAFU caused by it being planned by people who had no technical insight. Similar efforts in other parts of the world have been more successful.

Yes the idea is good but the implementation is lacking but there is still a strong push for it. I am also aware of OSS products that they have been implementing(you can see them in the tenders) but at the end of the day OSS needs internal resources that are capable of maintaining the systems and currently I'm not aware of many government organizations that have those skills. In my mind FOSS is perfect for government but the skills need to exist as mamongato and others have mentioned to maintain and enhance those products.

 

[Beachless]

The failures with the SITA may have been caused by 5 year performace based contract thing, or perhaps realised there wont be enough "kick-back" money from suppliers once they go FOSS.

I think its probably easier to hide kickbacks in services than product sales, to me the problem lies in that FOSS requires better planning, management and more effort its much easier to just spend big once off and get a working product and lean on the vendor/manufacturer than to implement FOSS solutions.

 

[mamongato]

What type management skills one need to implement it. I mean to me I don't think is something that difficult, a normal change procedure must just be followed isn't it?

 

[Beachless]

What type management skills one need to implement it. I mean to me I don't think is something that difficult, a normal change procedure must just be followed isn't it?

Its not as simple as it sounds for an organization of that size you must remember its not just a case of doing word documents and email there are much more complex applications, custom applications and back end services that are running.

The migration has to take every scenario into account and if a small percentage of the users are unable to do their job you are in trouble. Read through the docs I linked to earlier to see their plan it lists some issues and risks and should give you some insight:
http://www.sita.co.za/FOSS/Draft FOSS RoadmapStrategy.pdf

 

[Zyxoas]

How will they be more secure? Closing loopholes as they happens is hardly security, that means it's wide open until someone finds the loophole, it could take years. In this case the larger the username the faster problems are found and patches....therefore chances are better that larger and possibly commercial options are more secure by default.

That's a bizare thing to say...

The reality is completely the opposite. OSS implies many eyes looking at the source and monitoring repo commits, which therefore results in security holes being caught quicker.

FOSS shops tend to have a "release often" philosophy. You do not need to wait several months and jump through flaming hoops to have a zero-day bug fixed -- you simply "sudo apt-get update" (on Ubuntu) and all known vulnerabilities in standard packages are fixed.

Also, you seem to be assuming that FOSS = niche. Android runs a third of all smartphones, and is FOSS. Apache, nginx, and PHP run a large chunk of the internet, and they are all FOSS. Chrome is based off Chromium, which is FOSS.

Although you might have more eyes looking at IE6 than Chrome, you certainly have a lot more users using Chromium technology as opposed to Microsoft's latest half-baked attempt at a browser, and with a large community of developers reviewing and contributing code, and a fast-paced release schedule (Chrome is 35.0.1916.114 after less than 5 years in the field) implies that bugs will be caught and fixed far quicker.

 

[SauRoNZA]

That's a bizare thing to say...

The reality is completely the opposite. OSS implies many eyes looking at the source and monitoring repo commits, which therefore results in security holes being caught quicker.

FOSS shops tend to have a "release often" philosophy. You do not need to wait several months and jump through flaming hoops to have a zero-day bug fixed -- you simply "sudo apt-get update" (on Ubuntu) and all known vulnerabilities in standard packages are fixed.

Also, you seem to be assuming that FOSS = niche. Android runs a third of all smartphones, and is FOSS. Apache, nginx, and PHP run a large chunk of the internet, and they are all FOSS. Chrome is based off Chromium, which is FOSS.

Although you might have more eyes looking at IE6 than Chrome, you certainly have a lot more users using Chromium technology as opposed to Microsoft's latest half-baked attempt at a browser, and with a large community of developers reviewing and contributing code, and a fast-paced release schedule (Chrome is 35.0.1916.114 after less than 5 years in the field) implies that bugs will be caught and fixed far quicker.

My point was that it isn't simply a hard and fast rule of FOSS = security or commercial being the opposite.

Commercial products can be as secure as FOSS products, just like FOSS products can be equally as insecure as commercial products.

There's no universal law that applies.


You blow a horn about Chrome being on version 35 in less than five years. I'll argue that BECAUSE of major version changes and too regular updates there's a bigger chance of introducing new security holes and/or bugs, whereas the likes of IE retain stability with more infrequent updates.

I'm not saying the above is the case, I'm simply making an argument against yours.

There is no universal rule here like the OP seems to allude to, it's purely a case by case basis.

FOSS is cheap to implement in small environments but often not the case in larger ones. Not when all aspects are considered.

*****

I don't want to make it sound like I'm all for commercial and have something against FOSS.

I'm running an 85% FOSS shop here, I'm just stating that the OP isn't correct in their assumptions and haven't considered all factors.

 

[mamongato]

FOSS is cheap to implement in small environments but often not the case in larger ones. Not when all aspects are considered.

National Library of SA is large envrironment, it was implemented cheaply and they are still reaping the benefits. Same technicians retained to work there even after migration from commercial. 500+ servers and desktops in PTA and 60+- in CPT.

 

[w1z4rd]

Not true, switching software means the company will have to train staff, also there's more places offering Office support not that much people need it.

The open source office applications have basically the same layout as the paid for versions. So no.