I love Linux, and love Windows too (especially for games). It does make me happy to hear about Linux deployments like this.
That said: Everyone who switches one way or the other claims savings. Their breakdown of savings raises questions. They have 15,000 machines and TFA says they'd be paying €280 a seat for Office upgrades and €173 a seat for Windows upgrades. I have worked for Windows companies with less seats, in the commercial space, that paid less.
I also want to know what the workers think. I've tried OpenOffice but find the spreadsheet unusable compared to Excel. My low productivity in OpenOffice makes Office a bargain.
Power Excel users use the keys (rather than the mouse). The new Office button bar forces you to loose all spacial understanding of where obscure menu items lie - fortunately, Microsoft retained key-compatibility with previous versions (even though those keystrokes no longer make any sense w.r.t. the visual interface).
The debate that Microsoft has on cost per seat is dwarfed by retraining costs - which may actually be lower for LibreOffice than the transition onto the ButtonBar in Office 2010. Of course, Windows is going to move to Windows 8, and Linux desktops will be more familiar than ever...
Personally, I've moved over to LibreOffice, writing a Python addin to get Bloomberg compatibility. My users are on LibreOffice for word-processing (practically no-one has noticed), for spreadsheets I'll let them flip back to Microsoft Excel if they have 'power user' problems/mindsets - but often find that they are taken aback by the button strip anyway.
I didnt't seriously start using Excel until the 2007 version, but I must say that the Ribbon is much better than the old system. The discoverability of keyboard shortcuts are much higher when you only have to press the alt key to reveal them.
Of course I also use some other shortcuts that are not on the ribbon, but that's just a bonus.
They aren't shortcuts, they're replacements for mouse clicks, and you have to do the same number of key presses as mouse clicks.
I don't have Windows open in front of me (still the holiday weekend). In Word, to paste text from the clipboard, unformatted, into the document by "pressing the alt key to reveal" the so-called shortcuts go something like this:
- alt
- H (I think, to further reveal or expose the Home "shortcuts")
- V (to expose the paste functions)
- S (because we want to paste Special)
- U (unformatted)
- and maybe U again to get the other flavor of Unformatted.
That's five or six keystrokes, and that's not a shortcut. It feels like passive aggressiveness on Microsoft's part: "Oh, you don't like clicking? ... There, I fixed it."
Excel 2010 is certainly more novice friendly, but I'd take 2003's interface back in a flash - customising/grouping shortcut icons doesn't seem nearly as friendly in 2010 and it's a big drawback. I also find it much quicker to scan lists of vertical text when looking for a menu item than to scan horiztonal lists of icons/text.
As an aside, I'm guessing the Python add-in you mention is just wrapping the C SDK. You'd probably be interested in the Python SDK which is now maintained by the API team as part of BLPAPI v3.5 :)
a purely business economic point of view does not make sense in public spendind. No highway is econimical for itself. One must consider the national economic benefits. With free software, it is clearly higher than for proprietary. The license fees fund the stakeholders of an American monopoly. Expenditure on migration, training, adaptations benefits local companies and people, improves and promotes the local know-how and the common good free software.
This is the kind of thinking that makes the public sector a quagmire of inefficiency. The IT department of the city of Munich has exactly one mission and that is to enable the employees of the city to service the citizens of Munich in the most efficient manner - not playing politics on behalf of the free software movement (or Microsoft, for that matter).
By the same yardstick, the city of Seattle should stick with Microsoft everything to protect local jobs, in spite of evidence of superior alternatives.
"The IT department of the city of Munich has exactly one mission and that is to enable the employees of the city to service the citizens of Munich in the most efficient manner"
Exactly. It is more efficient for Munich as a City to support local business than to transfer money to american sharehoulders, if they think faresighted instead of shoresighted. Because part of the money comes back as taxes.
National economy is not a profit business but a money circuit. Therefore the efficienty laws are different.
The goal max. efficiency is the same, but the ways to accomplish this are totallly different, because the economic systems are different.
The city shouldn't maximise for profit, that's clear. But in this case they might be optimizing for a vague political goal at the cost of quality of service. Stuff like, a planning permission taking longer to process, but, hey, we benefited the local open source community.
I'm not saying that's what happened, I'm saying that's the kind of unintended consequences you risk running into when you muddle the mission goal.
By the way, any contract of any meaningful size must be tendered EU-wide. It's illegal for a city to demand that its providers are local.
The quality of service in Munich is excellent.
The goal is not vague but a very clear strategic one. Be not dependent on a foreign monopoly. Support and adjucate your own people. That is the constitutional duty of any government. And Munich excells in this regard with the help of Free Software.
But using Linux was not a decision of the IT department - it was a political decision by the city council and the mayor. If they decide to give a subsidy to Linux to promote local software development, that is perfectly within their mandate.
By the way, you need to realize that Munich's finances are very sound. In 2012 the city had a budget surplus of around 700 million € - so spending 20-30 million € on a 10 year IT project is not something that hurts them very much...
The council playing politics with the IT department is the same thing. Yes, it's within their mandate, but it's still the kind of decisions that makes public services so crappy.
The only source of income for the city is its citizens. The fact that they have apparently been overcharged to the tune of €700 million is not a license to squander €20-30 million - on a project that might decrease the quality of service.
I've only visited Munich once, but from my experiences there of everything from the police to transportation or keeping the streets clean it seems to be one of the most "got their shit together" cities I've visited.
I challenge you to read my original comment and the comment I replied to. Seriously.
I'm challenging the idea that there is a difference in how you measure the success of a public sector entity over a private sector one, not wether Munich is a competently run city.
In fact, the OP suggested that is was meaningless for Munich to have done this private-sector-style savings analysis. I wholeheartedly disagree with him and agree with the approach of the City of Munich.
"The IT department of the city of Munich has exactly one mission and that is to enable the employees of the city to service the citizens of Munich in the most efficient manner"
Does the City have only this purpose? Is there not a secondary purpose in stimulating local business/spend?
In the UK we have local councils making people redundant and replacing local services and contact with call centres in a distant timezone. The cost saving by doing this is possibly offset by the increase in social security/benefit bill (from a national budget of course).
I was present in a hearing by the Bundestag (German Parliament) of the director of LiMux project.
I particularly remember two of his points:
The main goal of the project was to reduce vendor lock-in, not financial savings. It is unacceptable for a governmental entity to be so dependent on a single corporation (especially if the majority of that company doesn't fall under the countrie's control).
The main problem of the migration was that so many third parties insist on exchanging data through office documents (even though they are often ill suited for that), instead of providing an API or a simple data file format.
I wonder how the quality of their Linux systems compares to an equivalent Windows set-up. Do they have more crashes? Is it harder to train people to use Linux? More problems with program compatibility? Is Linux as easy to use, or are there inefficiencies? Saving money is one thing, but time = money so if using Linux wastes time then is it really an advantage?
Maybe I'm just looking for problems. I'm a bit biased; I happen to think it's a matter of self-respect to spring for quality software instead of just always using the cheapest option. It's one reason why I would prefer to use MS Office over OpenOffice, because Office 2010 is a quality app with a polished interface whereas OpenOffice and its variants look to me pretty much like crap on a stick. I also like the idea of rewarding quality software and promoting more of its kind with your money. We should all be happy when smart people making good products get paychecks.
However, like you, I'm a bit biased too - except the opposite way. To me, some of the questions you raise (Do they have more crashes? [..] More problems with program compatibility?) are points I would traditionally associate with Windows rather than with Linux.
Is it harder to train people to use Linux? is an interesting question. I think if people are already used to working with MS products staying with these products should provide an advantage it terms of training. However, I would expect a significant portion of the staff to use only a minimal feature set available in their setup, typically writing letters, etc. For them, I wouldn't expect the transition from MS Office to an open source office suite any harder than, say, the transition from traditional MS Office to the ribbon-based interface.
Is Linux as easy to use, or are there inefficiencies? -- The whole "this year is the year of Linux on the desktop" joke will probably never get old, but I think everyone has to admit that Linux has come a long way in the last 15 years. For your average user, I would say there aren't any major differences between a Linux desktop and that of a commercial vendor, once the system has been set up. Of course, that still leaves the question whether Linux systems are set up as efficiently as Windows systems...
I personally think that MS Office at least looks way more polished than any of the open source alternatives, and I would expect it to be a good deal ahead of the latters in application as well. But I can't say for sure because I've used Open-/LibreOffice exclusively for many years now. It's totally fine for whatever I want to do with it, and I would bet that it is absolutely sufficient for 90% of the use cases an average admin person has.
So all in all MS products might still be a (good) notch ahead in terms of product quality, but I think that advantage is somewhat neglectable because Linux alternatives have certainly matured enough for the particular use case of a city administration.
I found it felt very natural to use Linux Mint coming from Windows 7. In fact I even thought Mint is more intuitive than Windows 7. I can't say I had the same experience with Ubuntu's Unity, though.
Also I need to try out Zorin OS (zorin-os.com) one of these days. It looks like they made it so it works as much as possible like Windows 7, so it should be even easier for people switching from Windows.
Note that they were using OpenOffice, not a more modern fork, and that they were running it in parallel with MS Office 2000, so using some form of Windows.
They're in for a rude UI awakening when jumping to MS Office 2007 or later...
Yeah, that's a sad example of what vendor lock-in closed file formats cause and what bad migration planing does. I guess IT people in Freiburg could learn a lot from their fwllows in Munich.
Personally, I like the contrast: Freiburg will get to experience Windows 8 and update us on that, while Munich will update us on how much money they saved. That's what this article is about: how much money they saved.
The hardware update savings struck me as a bit contrived. Maybe they didn't technically need to upgrade hardware in connection with OS upgrades, but hardware needs to be updated periodically regardless. Seems like they are just disassociating the timing and claiming savings.
I'm not sure whether you were aware that almost every major Windows release has required significant new hardware purchases? Corporate IT usually delays rolling out each new Windows (preferable, but senior management gets impatient), but that just spreads the increased TCO over the hardware refresh cycle.
2006: Vista premium required new hardware because it required WDDM1.0, and the first cards to support it were the NVidia 8800's [1] [2] and ATI R520 core [3]. Note the whole "Vista capable" fiasco [1] [5], hence Vista premium is the relevant point.
Windows 7 hardware requirements were the same as Windows Vista.
2012: Windows 8 requires at least 8 GB of RAM [4] even though Microsoft claims it has the same requirements as Vista (1 GB RAM). That also means the 32-bit version of 8 is crippled from the start. UEFI Secure Boot, also "not a requirement," will increase the hardware TCO for IT departments purchasing machines meant to have Windows 7 – they first have to disable Secure Boot.
It's understandable that machines from 2006 will not still be around in 2012, but Microsoft is definitely the cause of the increased hardware TCO.
Easy there, tiger. I just upgraded a Dell Latitude E6500 with 4Gb of RAM and an Athlon 5050e desktop also with 4 Gb from XP to Windows 8 Pro. Anecdotally, both boot faster and run Office 2010 faster than when they were running XP. This major upgrade has required no new hardware, except for a Logitech QuickCam Pro for which there is no 64-bit driver. I can probably run the desktop for another 5 years now.
You can't be serious, you are of course joking, and if you aren't then you are insane, and you reference an opinion piece as your source for that? Wow.
2012: Windows 8 requires at least 8 GB of RAM [4] even though Microsoft claims it has the same requirements as Vista (1 GB RAM). That also means the 32-bit version of 8 is crippled from the start.
I use Windows 8 and Visual Studio 2012 in a 4GB virtual machine. This works fine and is fast.
Contrived is a bit uncharitable. 5 year old hardware (which would be around the calibre of a Core Duo with 1Gb ram?) is still perfectly serviceable for your average office worker.
With Linux they have the choice to defer hw upgrade until it's justified for reasons other than 'we need to upgrade the OS and because the current hardware came with an OEM licence and so is not covered by any volume licencing deal that may be in place (without spending extra) its cheaper to buy a new PC with a Windows bundled than it is to just buy the software upgrade then the hardware when it's really needed'.
A city like Munich would, I would think, have a volume license arrangement and not rely on OEM licensing; they'd buy the hardware sans OS and provision their locally-configured OS image. I've never had issues with installing newer Windows versions on older hardware (within reason) in this scenario; though I don't use Windows as my primary OS I do have it around for when I need it.
If you were using Linux as your main OS, you would have also the same opinion.
When you have a fast machine with windows, 3 years later, it is slow as hell.
When you have a fast machine with linux, 3 years later (in fact 4 years in my case), it is (almost) the same speed and you do not want to change the hardware because you are afraid of potential hardware issues.
Companies and public services who use commercial closed software usually pays someone for creating it - will they now send money upstreams to the different software projects that makes it possible? GNU, Linux foundation, Document Foundation, GNOME, KDE and Mozilla?
I spent the summer working with a lot of old time programmers who were in the field since the early 70s. To me Linux was a privilege to them it was stingy: they wanted PowerPoint, Outlook and did their work in Putty.
I brought this up previously[0] in a discussion where Helsinki didn't go with LibreOffice as they found it would have been $21M Euros more expensive than staying with Microsoft. I'm reproducing my bullet points here as I think the claimed savings in this situation are actually false savings. They supposedly took training into consideration but I think they ignore loss of productivity inherent in the switch as well as compatibility issues. Consider the total costs of all the following:
* How will incompatibilities be handled with their existing documents? Is LibreOffice absolutely 100% feature complete identical to Microsoft Office? If not, what proportion of documents use incompatible features? How much effort is required to determine this incompatibility? How much effort is required to redo these documents to fix the incompatibility? How many employees are required to be on such a conversion team and how long will the conversion take? Can the conversion even be done?
* How is interaction done with external users who send Office documents? What happens when users outside their office send them Office documents which make use of incompatible features? Will this require maintaining an ongoing conversion team? What is the opportunity cost inherent in the delays in getting documents converted?
* How is installation of LibreOffice done? Can it be setup on a centralized WSUS server and automatically deployed to every user as Microsoft Office can, or does it require an army of people manually installing it on thousands of PCs? How are updates for both version upgrades and security patches handled? What is the typical update schedule for LibreOffice (security-related patches, compatibility upgrades, etc.) and will this need to be done manually as well? How many man-hours per year will this take?
* LibreOffice doesn't include an email product. What alternate solution would they use? What is the cost to convert all Outlook email and archives to this new solution? Can the new solution make use of all calendaring and other Exchange-related features that Outlook has? Can the new solution be deployed on WSUS and kept up to date automatically, or does this too require an army of people manually installing and doing updates? Can they maintain the security permissions they currently use, such as restricting the ability for users to forward internal communications outside the office?
* LibreOffice doesn't have a Visio, Project, or OneNote replacements. What are they going to use for these? How do they convert existing documents and interact with external users who send them Office documents? Is it even possible?
Sorry, but a lot of these points look like Microsoft-playbook FUD. I'm guessing that you're trying to get more hits via SEO on a hacker-friendly website.
However, most of the people here are Linux by default on their servers, and whatever comes natively on their laptop - which for most web developers seems to be OSX these days.
To address your points (quickly) :
Incompatibilities : DOCX isn't even compatable between Microsoft products, and we recently noticed that gmail converts it silently to DOC : Making it a broken format if the documents are ever sent over the web to a client.
Interaction : Not as much of a problem as Microsoft would like. Clients seem quite ready to accept formatting snafus as Microsoft's problem. For WYSIWYG, use PDF. For editing, formatting is flexible. Conversion team HA!
Installation : LibreOffice on Linux is a much more scriptable solution that Windows/Office. Purely because of licensing issues. Difficult to take a lecture on security...
Email : Selecting a decent email client is pretty easy. Particularly since Microsoft Outlook's implementation of IMAP is so broken. If you're interested in restriction employees from forwarding emails, I guess that's something that could be handled by some sort of corporate email relay - which will probably run quite nicely on a UNIX system.
Visio, Project, or OneNote : Actually, not everything has to be bought from Microsoft, does it? This is kind of scraping the barrel, unless your solution requires all software to be bought from Microsoft.
I'm pretty certain you'll have plausible-sounding rebuttals to all these points. But (short of downvoting me off the page) my off-the-cuff response will be linked to your copy-paste effort for everyone to see.
Sorry, but a lot of these points look like Microsoft-playbook FUD. I'm guessing that you're trying to get more hits via SEO on a hacker-friendly website.
Your post was great aside from this indulgence of paranoia. (Well, that and the denial of OneNote, because even though I'm a 100% Mac user at this point I still miss OneNote; there really is no alternative and I'm still sad about it.)
And I did downvote you because of it, though IIRC it gets wiped because I chose to respond to you instead. Accusations of FUD and the "Microsoft playbook" aren't right.
If I was wrong in suggesting Astroturfing, then I apologize. And thanks for un-downvoting me.
Looking at what developers are choosing (when given a free choice) also makes me think that Microsoft is losing ground. They used to be able to claim "It's better" - a much more enticing sales pitch than "You can never leave".
On the enterprise front, going forward, it's going to be easier to get management interested in web/intranet delivered solutions via the PC (or iPad) browser. Far fewer worries about roll-out, hardware upgrades, etc.
One of the big arguments that Microsoft (and its partners) has always been switching/training costs. And with the Office 2010 button bar, coupled with Windows 8 'tiles', I think they've created a massive own-goal. It's pretty easy to roll a rock-solid Fedora/XFCE/LibreOffice/Thunderbird/Firefox image, which will feel very familiar indeed to users.
I wanna know why people use MS Office, anyway: 99% of the communications are: Please click this link to open this Word doc on a web server.... couldn't we just post up a HTML page instead? I bang my head at work every time someone does the above. It's a big waste of time, and money. And don't get me started on SharePig
I agree with you, for the most part (though I actually do like the Ribbon when I need to use Office). I just find slinging around claims of astroturfing to be counterproductive, especially when your points are solid. :)
If a comment or an item is worth a downvote, it's not a valuable contribution, nor does it contribute to the conversation. Replying to it then becomes pointless, and you are replying to nothing of value and something that should not contribute to the conversation. This would, as a result, make your comment equally worthless and equally devoid of value.
By commenting, you are suggesting that the comment does contribute to the conversations as its assumed that you would try to make your comment worthwhile.
You can't have it both ways. Either the comment is worthless and should not be read or replied to, or comment contributes to the conversation by at least sparking interesting and worthwhile followup commentary.
It doesn't need to have 100% of Microsoft's Office's features. A large portion of them are most likely unused by most Government employees. For e-mail clients and whatnot they can find open source alternatives like Mozilla's Thunderbird. LibreOffice doesn't need to have an e-mail client.
And what you're talking about in the beginning may be true if say the alternative is to simply stay with Microsoft right now versus switching to Linux/LibreOffice. Then yes, obviously doing something costs more than doing nothing, although the savings may still be worth it over the long term.
However, if the question is whether to upgrade to Windows 7/8 and Office 2010/2013, then it's very likely the Microsoft option costs more. Plus you get the added "benefit" of being locked in over the long term, and spend a lot more money continuing to invest in Microsoft products, which includes training of new employees with Microsoft products anyway.
Most of office is pretty replaceable, but the two bits that aren't are Outlook and OneNote. OneNote is semi-niche so it's not as pressing, but I would strongly disagree with the notion that Thunderbird is a replacement good for Outlook.
Seeing what practiced Outlook users can do is really eye-opening; it's a tool a lot of nerds (including me) hate, but it is a really powerful information management tool and I don't think Thunderbird (or anything else, including Evolution). I don't think I've found anything that is, though the full Google Apps suite does a decent enough job for my personal, non-work purposes.
Ex outlook power user here. Sorry but outlook is almost certainly used wrongly by most people. Its used as a filing system rather than a communication tool.
Mutt is just as useful if you don't mind not getting cat pictures.
We have a public JIRA instance. Everyone uses it or gets told to piss off. An email results in either a ticket being raised and is deleted. inbox zero. It handles the entire organisation's processes and rules and most importantly audit trail.
Scheduling is handled by an ical server which has feeds from jira. People use ical, their phone or mozilla's calendar software which I can't remember the name of.
Documents live in svn.
All of this is backed with centralised authentication via active directory/ldap.
There is nothing else. Outlook is the wrong tool for most jobs.
You do not get to tell users that they use a tool wrongly if it effectively serves their purposes. It's great that a Jira-based solution works for you, but that does not make it "right"--it makes it work for you.
Your attitude is why normal people hate IT people.
Hell yes you do if it doesn't comply with the audit, security and distribution requirements of the organisation.
Outlook effectively shoots all of the above without massive customisation which is expensive, ineffective and has so many edge cases it's unreal.
People hate IT people because they don't like being told what to do and don't like learning how to do stuff which is outside their comfort (read incompetence) zone. When it comes to handling your company's data, doing what you are told to do and following the process is the law.
The integration is what sells it, for sure. Between Exchange and Sharepoint, you have a really nice solution for normal/non-techie workflows. I hate it, but I also am pretty comfortable with alternatives and do most of my own stuff in a terminal and browser.
That said: Everyone who switches one way or the other claims savings. Their breakdown of savings raises questions. They have 15,000 machines and TFA says they'd be paying €280 a seat for Office upgrades and €173 a seat for Windows upgrades. I have worked for Windows companies with less seats, in the commercial space, that paid less.
I also want to know what the workers think. I've tried OpenOffice but find the spreadsheet unusable compared to Excel. My low productivity in OpenOffice makes Office a bargain.