Slashdot Mirror
Convincing Your Employer To Go With FOSS?
mark72005 writes "My employer is currently looking at adopting a content management system for use by our technical support staff (primarily first-line end user support, but hopefully it will include deeper levels of support personnel eventually). The candidates are currently Plone (OSS) and Confluence (proprietary, closed-source). For those with experience in each, what arguments in favor of Plone could be made to managers more interested in pragmatism than idealism?"
Cost?
My problem has been convincing them that they con't just pass of the cost of Windows to the customer. They like the fact that they can hire 3-4 MCSEs for the cost of one good Unix admin, but they don't realize that the Unix admin can set things up so that maintenance is much easier.
Windows is ingrained in business culture here, for the most part.
Free. Thats really all that is required here, but then I work for a bunch of cheapskates who won't be around much longer.
"I use a Mac because I'm just better than you are."
I've used TWiki (OSS, all Perl IIRC and aimed at corporate usage) at one job and Confluence at another but not Plone. Confluence is good for non-technical people because it has a pretty good wysiwig editor, but its search was simply wretched. I think we had a lot of 'lost' knowledge in the Confluence DB because nobody knew it was there and the obvious searches didn't show it - I would come across nuggets now and then. If you have the discipline to build index pages, it's probably a good choice if you have a lot of non-engineer type people.
TWiki (and this was a number of years ago so it may have improved) was almost the reverse. Good search, good architecture for plugins, but no wysiwyg so non-technical contributors had trouble with it. They were writing a wysiwyg plugin so that may have now arrived. It was easy to maintain and of the two I would say I like it better.
Graham
I once asked Richard Stallman how to convince my school to go with FOSS instead of Windows, since most of our CS lab was on Windows.
His reply: "Defenestration! Throw Windows out of the computer, or throw the computer out the window!"
There's no -1 for "I don't get it."
Might I suggest Liferay (http://www.liferay.com)? Open source, but also commercial, and more featureful than both Plone and Confluence.
You've obviously decided which piece of software you want to recommend even though the only reason you can think of to recommend it is that it is FOSS? If the open software isn't as good it just isn't as good; just because it's FOSS doesn't mean that it is the be all and end all to solve your problems. Compare features, stability, cost, and support; if your boss is actively against FOSS make a point to explain it's advantages (and disadvantages if you want to be fair) and leave the decision to him. After all, it's entirely possible that the closed, proprietary solution fits your situation better; basically, its dishonest to make your decision and then go digging specifically for evidence to support that decision.
Protection against lock-in is something employers understand the value of.
This is my signature. There are many like it, but this one is mine.
I speak only from my works own dynamics - If opensource software was to appear on work machines(lets say an open office variant) it would last as long as one of our managers receiving a docx from some outside manager with fancy things(annotations, drawings) and the ensuing discussions as they work out they are not looking at the same thing. The manifestation for me it the manager would turn up at my desk with his "look of death" and the question would begin "Can you tell me why...." Been there, done that. The whole thing falls like a deck of cards.
In post Patriot Act America, the library books scan you.
http://moinmo.in/
I haven't used either system but the podcast FLOSS weekly recently did a whole episode about PLONE that may help you decide if it is right you.
"UNIX is very simple, it just needs a genius to understand its simplicity." -Dennis Ritchie
Confluence integrates with Jira. I like and can't argue against it.
I've never used Plone, but as the old cliché goes, best tool for the job.
Our experience with it (large internal reference system at a major consumer product business) has been terrible.
Don't worry about FOSS versus proprietary (at least as a sole arbiter)...worry about finding a good system and all other things being equal, choose the FOSS one...but whatever you do don't choose plone.
One cool thing about F/OSS is you can simply install a working version - get it up and running.
In most companies, from that point momentum alone will keep it.
That's why we're using Postgresql/postgis instead of Oracle/ArcSDE.
That's why we're using Lucene/Solr instead of FAST and the other ones we were supposed to evaluate.
Just tell management that you're setting it up as a proof-of-concept; and that the various commercial vendors are welcome to install theirs as well; and you'll stick with whichever works best.
Speaking as someone who has worked with Plone, I recommend you don't use it.
Atlassian makes the source for all of their products available to anybody who buys a license. It doesn't cost anything extra, and even the $10 starter licenses come with full source.
Perhaps the question you should be asking is, cost aside, which would better suit your needs? Sure FOSS is great but if there is a better fit for your needs and someone else is going to foot the bill, who are you to say that management is looking in the wrong direction? I, for one, believe that there is a place for both commercial and FOSS in the business (and in the home for that matter). Perhaps a cost-benefit analysis needs to be done. Ultimately the decision needs to suit the needs of the business and not the ideals of the employees.
.sig
Plone is a CMS, Confluence is a wiki. Incedentally, both products are quite good. I used Confluence at a previous job and it is a very nice wiki. We used it because of it's tight integration with Jira, an issue tracking system by the same software vendor.
Did you mount a military-grade, variable-focus MASER on an unlicensed artificial intelligence?
So are you saying you would rather have your boss make the idealistic decision? When it comes to business software, pragmatism reigns. It's the responsible thing to do.
The requirements sound like you need knowledge management system. Not a content management system.
Of course that being said (without knowing the requirements), why wouldn't a wiki work? There are lots of wiki solutions available.
If you compare features, Plone easily wins.
AFAIK Confluence is open source but not free, I recall Atlassian providing source code with their enterprise licences. I'm completely unfamiliar with Plone, but I've investigated a fair number of Wiki's and CMS's. From what I learned a while ago when looking into this, there is a distinction between a wiki and CMS, with CMS being a bit more complex. One of the basic CMS requirements was workflow publishing, i.e. content is changed but not made available for public consumption until the necessary parties had reviewed and authorised it. I was very impressed with XWiki and Alfresco
RTFM is not a radio station.
Tell that to all the managers who choose to buy MS Office just for word and excel in this day&age, or who insist on windows for terminals that do nothing more then browse the web with firefox to get to a single web application.
This gets much less attention than it deserves:
http://www.fsf.org/working-together/whos-using-free-software
Testimonies from Cern, NYSE, the EU, Wikipedia, and the US Department of Defense, plus another page of testimonies from individuals:
http://www.fsf.org/working-together/profiles/meet-the-free-software-community
Expert in software patents or patent law? Contribute to the ESP wiki!
I know this isn't one of your stated options, but ModX 2.0 is worth a look. It's so well optimized for SEO that we cloned our site in it as a test, switched it on, and within a week its organic search ranking was just under the original that we pay $40K/mo. in paid search to promote.
Do what you can, with what you have, where you are.
Keep it simple stupid - there is the old saying that no one ever got fired for buying Microsoft.
What never gets added is that people have gotten fired for going above and beyond to advocate for FOSS and then got fired when there was a show-stopping problem (which can happen no matter what new scheme you bring in).
FOSS has it's time and place, but *you* sticking your neck out trying to jam in FOSS into an environment that is not culturally ready for it is just asking for being the center of a CYA shitstorm.
I'm guessing that a bunch of people on slashdot have been severely stung from drinking the kool-aid. It hurts the company, the boob that was a advocate instead of a advisor, and most of all it hurts FOSS.
Don't push, FOSS will get there on it's own schedule.
Humor from a Genetically Molested Mind
If you want to make a solid business case, you need to approach it objectively; what option will cost the least, in the short, medium and long term?
Maybe it's OSS, maybe it's not. But drop your bias right now before you research associated costs.
Mod me down with all of your hatred and your journey towards the dark side will be complete!
I'm a huge open source advocate, but Confluence (and JIRA!) are very hard to beat. If my employer was buying either Confluence or Jira, I wouldn't fight over this *at all.*
-fb Everything not expressly forbidden is now mandatory.
I have been all up and down that road. Cost isn't the only consideration. Accountability isn't the only consideration. Perception is often the most important consideration. Think of the difference between an H2 and a really nice pickup truck. If you think an H2 is a Humvee, you would be wrong! The construction of an H2 is a lot like a pickup truck. But the perception that the H2 is a civilian Humvee remains. So, people kept buying those stupid, over-priced pickup trucks thinking they were something they were not.
People also think that commercial software comes with "accountability behind it." Well, that depends on your contract with the vendor and if anyone has read an EULA, they will tell you that they make no guarantees about suitability or applicability or reliability. They take no responsibility for data loss or any loss at all resulting from the use of their products... it goes on and on and on... you buy it, you get to use it, but if anything bad happens, they will be happy to send you a copy of the EULA with yellow highlighter indicating the relevant disclaimers. But somehow, business people have it in their heads that "they have someone to sue if things go bad." (All that said, some sate laws do enable accountability despite the EULA... "void where prohobited.")
And as far as cost goes? "ROI" doesn't calculate so well when the I = 0. It confuses people. If you want F/OSS in your company, get a vendor to sell it to you as a support contract.
Lock in can be caused by a software writer no-longer being interested or able to support or extend the product. Yes it can be because the company decided it should die. But, in the OSS world, the practical equivalent for a complex piece of software is for the original developer(s) to quit the project and no one pick it up. Is your company going to fund the time for you to learn to maintain the product? What will happen when you leave?
I once won a lawsuit based on the complainant claiming they could maintain the complex code as well as the original writers of the software. I said of course that fixing was theoretically possible but could they do it as effectively, delivering the fix in as short a time and as reliably? Yes your honor, the answer is somewhere in your law library, now go find it and figure how how it all works together. The original developer or one who has studied it just has all that how it all works and where to start looking in his head (or her head, or their collective heads).
Complex stuff is complex and part of the selection process should be an assessment of the financial stability of the provider and their reputation for support or (for OSS) the number of contributors who are fully knowledgeable and thus are protection against the hero developer getting hit by a car tomorrow.
Not to mention documentation, training resources availability, migration support to move you from where you are and even to migrate you off the product you are now planning to use. Its all that heavy duty life cycle stuff some products have and some don't and won't.
Think in terms of how this investment is going to pay off and how it will increase productivity and reduce risk. When you argue that way and provide a features comparison as an afterthought you can successfully advocate a solution to a business problem, until then you come off as a geek who doesn't understand the business problem they are trying to solve ... which is to produce more profit.
Actually, I'd go with Confluence. It's not OSS, but it's and awesome Wiki. Choose what's the best tool for the job, not what suits your religion.
It's baffling that the question is posed that way so often. For a rational businessperson, the question is really:
What argument could be made in favor of paying for a software package when one of equal or greater value can be had for free?"
1. Written in Java, which means you're more likely to have on-site language expertise in case something goes seriously awry (you get the source when you buy a license).
2. Lots of support available, as it's the most popular enterprise wiki system.
3. Integrates with SharePoint, which for many places is a must-have.
Basically, Atlassian focuses on the enterprise market, and it shows. Best tool for the job, etc.
I've used Plone as a CMS in a company before and here's what I can tell you.
Plone security works great especially if you fine tune it. For example, you are definitely going to want to think about going in and tweaking what happens when documents move to different publishing states. I tweaked the "Publish External" to have the same privileges as internal publishing because for us, there was no such thing as external publishing since it was an internet facing company intranet and client extranet.
You will also want to proxy your access behind Apache if this is going to be internet facing.
Plone has a great ability to version files. Unless, of course, they are large files. IIRC, anything greater than 32MB causes versioning to fail. I know you can get around this by using external storage (external to the PloneDB) and I think they made it easier with version 4 that was just released, but I haven't tried Plone 4.
Plone is written in python, so if you want to build your own plugins, you are going to have to learn it. The built-in DB is like nothing I've ever seen and is not relational in any meaningful way that I saw, so if you ever have any ideas of doing something relational with it (i.e. a trouble ticketing system), you are going to have to use an external database for your plugin.
WebDAV works great in Plone. Versioning with it does not. Pick either versioning or WebDAV access for a folder.
Oh and unless things have changed, you cannot (AFAIK) do file level restores from backups. It is an all or nothing affair. You CAN restore to a test environment and then export an individual object to import on your live instance. For most issues of accidental deletion, you can recover from the management back-end though.
Like any solution, you will have lots of customization in front of you if what comes out of the box isn't sufficient for your needs. Depending on how dirty you want to get your hands with it, the learning curve can be gentle or very very steep.
Drupal has a huge userbase, and it is used by everyone (java.net is even run on drupal). Whitehouse.gov (and a bunch of other websites) use it. It has a low learning curve, it has a lot of modules to add functionality, it has a company behind it (Acquia) - so support isn't an issue, and there are plenty of people who know it.
Oh ya, and it is also free.
Speaking as a former developer for a major closed source CMS, I can tell you that open source is the way to go.
Here's how you SHOULD be approaching it:
1) Gather requirements from your key stakeholders - the people who will use it daily, the people who will administer the systems, and the people who will write the checks to fund the effort. (users, admins, managers) - define the use cases the tool needs to support.
2) Survey the available solutions and generate a list of the top solutions that appear to satisfy your requirements - this is your list of tools to investigate.
3) If you have time, do a hands-on proof of concept with each, or at least try to get some time poking around the live system, perhaps with a vendor demo.
4) Evaluate the PoC's against your list of requirements, and also pay attention to usability, maintainability, support costs, licensing costs, reliability, and fault tolerance concerns.
5) Compare how each one fares across all categories, and decide which gives you better value for your dollar - in other words, which is the better investment for your company.
6) Purchase a copy of that tool, and implement a pilot. Try to get the vendor to provide you with additional support during your pilot phase, or at least free licenses until the system goes live.
7) Work out the bugs in the pilot, then roll it out across your company.
If Plone can't stand on its own merits against Confluence, then you are sacrificing your own professional credibility and the best interests of your company and users to push an ideological agenda, and you will (rightfully) earn the derision and scorn of the people you are supposed to be supporting. If Plone is a solid piece of software - I'm not suggesting it isn't, I'm just not familiar with it - then it will probably emerge the winner in an evaluation.
And to all the people saying "just tell them it's free," since when is the cost of licensing a major component of the overall cost of supporting the software over its lifetime? The "free" system might be less stable and require a lot more (or higher-cost) support personnel to keep it running, and you need to take that into account if you're going to ask your company to invest a lot of money, time and effort into rolling out a new business system like this.
Having someone to call is very important. It means responsibility. FOSS sounds good, especially some popular stable projects. But do they fix bugs following your pace? Do they add features when you ask? Do they test everything before release or they let you test it? I personally work for a proprietary software company. I would suggest your company go with the proprietary software. Because you have someone to call where bugs happen! Well, bugs happen, and you will need to upgrade in the long run. Does your boss call you to work on it? If so, fine. If not, or you don't want to, go with proprietary.
The choice is never straight-forward. From a business perspective, it is often easier to go with a commercial solution rather than a stand-alone FOSS product for the same reason people rather invest in a hedge-fund rather some random high-yield bet: risk. If something breaks, there's someone else responsible for fixing it in a timely manner. It's also the reason Red Hat is able to make a business off free software.
The main things you want to look at when considering your options include: feature set (is one option missing features), support (does the commercial company have a good record for supporting their feature? does the FOSS solution come with some kind of paid support service?), and reliability of the software itself.
I used to spend the effort showing how several open source packages made things easier or solved problems. No more.
.NET product from a Microsoft gold partner their navigation bars wouldn't work in IE 8 but worked fine in Firefox). Save all these for the next discussion, and throw them in their face a bit. You have to bide your time when you see stupidity.
It's not just the problem if management thinks a product is inherently better if it's made by Microsoft, the problem is that no matter how small the change is, no matter how 99% of it works better, there's always one little thing that an executive with a chip on their shoulder can magnify, and proclaim to everyone is evidence that the "open source" product isn't better. No matter if this person is technically incapable of making such a decision, you get a C-level employee making noise and no one wants to argue with them, not even the other C-levels.
As a result, Firefox was discontinued, OpenOffice is in the process of being discontinued, and I don't even bother anymore. I'm happy they don't gripe about my use of Fedora Core on low-end fileservers and we're still a RHEL shop for our host systems (the bigger systems).
Sometimes you have to let them poke themselves in the eye a few times before they get it and are receptive. So I make sure they remember what a PIA "upgrading" users to Office 2007 has been, and how Office 2010 won't work with our Exchange 2000 Enterprise server, and how IE 8 doesn't let them open some business critical sites but IE 6 does (and yesterday when ordering a
Anon for obvious reasons
Have a look at Project Open http://project-open.org/ instead. It is a modular open source system that includes the mentioned components and more.
I've been trying to get my employer to switch from expensive Adobe Robohelp for years to some open source wiki software. We spend six figures on licensing every couple of years.
Their response is "well there's no guarantee that the software will continue to be updated."
For what Robohelp costs, we could keep an IT person on full-time who could customize the software to our needs and make adjustments or add features in realtime.
The problem with bigger companies is the same problem as objects with a lot of mass...it takes a lot more energy to change their direction than to change the direction of a smaller one.
You have two candidates and can install and test both with your particular use cases in mind in a matter of hours.
Liferay is better than both of those anyways.
If you think you can appeal to the guy on the basis of FOSS vs. proprietary software, then do it.
If you think there's a money angle (on initial cost, or on continued maintenance costs), then make it.
If there are things that Plone offers that Confluence does not, make a bullet list of those items.
If you're going to be the one maintaining this thing, spend a Saturday setting an install up in a VM so you can tell your boss all about how you already know how to use this tool. Human costs are often much greater than software costs, so if he thinks that the two offer roughly similar feature sets, your prior knowledge of the tool may tip the scales.
coding is life
You should be asking which of those two products best meet our needs of the users, meets the local it security requirements, and fits within the established budget.
personal preference and idealism is not the way to present a non-biased assessment.
It may even be the best option, but don't do it. I hate to say it because I work in IT and I do love a lot of FOSS but I have also learned to not push it in business. The only time I do recommend it is if the entire department/team is fluent in the OS/software/language being proposed... and in that case you wouldn't have to push to sell it.
Go with something that is supportable by everyone, that offers professional services, and fits the bill. Most likely that is not the FOSS option. I know it isn't what you want to hear, but it is the truth. I've been there, I understand, but I strongly advise you to reconsider. Good luck!
http://teasphere.wordpress.com - A little spot of tea
We use Confluence at work (as well as a bunch of other tools from Atlassian). It's ok, but it's search functionality has much to be desired (at least it's not as bad as JIRA's search functionality). What's wrong with using MediaWiki (the engine behind Wikipedia)?
OK, so I'm slightly biased in that I run a company that does predominantly Plone development, but one of our biggest clients uses Confluence to actually project manage the Plone project we are developing for them. So in effect we use both, for different sides of the same coin.
Confluence is quite powerful, and some of the tools for previewing MS Office documents in your browser are pretty good. However where Plone shines is its flexibility. We use Plone ourselves for our own intranet and the great thing is it goes well beyond just storing documents:
* It shows a list of all latest SVN commits from our repository
* There is a shared calendar with SMS alerts, iCal integration, etc
* It integrates with our time tracking software to produce time tracking reports for each project
* There is a wiki on there for ad-hoc knowledge bits
* There is a directory of all our contacts with click-to-dial integration with our desk phones
* All our quotes are 'written' in Plone and converted to PDF to send to the client. It handles all the formatting via Apache FOP
* There are image galleries for both social company photos, and also library of screenshots for quotes, etc.
* Management dashboard with graphs of time spent in the past week on which projects
So whilst Confluence is pretty good, I think you'd be hard pressed to customise it quite to the extent where it really starts to deliver business value by integrating with your actual business processes and other software.
Surely the right place to ask for help highlighting the selling points and strengths of Plone is on the Plone Forums?
You need features xyzpdq: Company 1 has features zyzpa, Company2 has features zyxpdb: Company2 also happens to be free software.
Yeah Me.
However, I suggest that you want the software that will help you do your job better. Crappy Helpdesk software won't help you retain knowledge. If employees who resolved a tricky problem leave, that knowledge is gone.
If you're making a decision for your business, you should be doing a feasibility study for both products, not asking slashdot how to convince your management one way or the other.
So the best way is this: do a feasibility study (i.e., operational, technical, economic, schedule research -- google it), present *your findings* (not what you feel about, then go from there.
Google Sites works well as a CMS
I've created sites with plone and use confluence almost daily at work. If you're after an easy to use wiki then you can't go wrong with confluence. It stable, has a lot of free add ons that really add functionality, and they release frequently with updates that really improve the product. Though the 4.0 release will give me pause due to the new unified editor.
Atlassian also seems to be a decent company (no I don't work for them). They give away licenses of confluence to open source projects, and their $10 starter packs proceeds go to charity.
If you can't or won't spend the money then I'd look at Xwiki. It's an OK substitute for confluence and open source. I use it at a site that doesn't use a wiki enough to justify the licensing costs and it works well.
My employer is biased to one way without good cause, and I am biased another way without good cause. How can I convince my employer that their bias is dumb, and my bias is better? I have no reason for believing the way I do, so please give me some. Thanks!
So the question is how you can convince your employer to go with shitty, open sourced rubbish over Confluence, a decent software. I totally see the reasoning behind this.
The fact that you company is exploring to significantly different products makes me think you may have bigger issues than just closed versus open source. Plone is a CMS system build on Zope, Confluence is a Wiki. I would not prefer to use Plone to capture shared team knowledge any more than I would choose Confluence to serve up my corporate web portal.
If you are looking to develop a customizable CMS system that will be used by both internal and external people with workflow support then lean towards Plone. If you are looking for a good place to easily capture shared content for your team then lean towards Confluence.
P.S. I have not used Plone much but have been a very happy Confluence user for about 5 years.
I registered here just to post this
:)
:)
Few years ago I was looking for a wiki solution for work. I tried Confluence, Socialtext, Dokuwiki, Twiki, Mediawiki and Deki. My findings through all that discovery was this:
Best user inferface and ease of use goes to Confluence, Socialtext and Deki.
The stupid obnoxious sales people from Socialtext and Deki quickly made me put them on the no go list. Confluence on the other hand was way better, I was not contacted by a sales person but by a friendly technician who just asked how my trial was going, if I had run into any problems she could help with and where to get support.
Twiki was amazing feature wise and the huge community behind it but it lacked polish and friendlyness. Also it was a pain to set up, documentation was of typical open sourceness, lots of detailed info but also lots of gaps.
I liked Mediawiki mainly for two reasons, very familiar interface for Wikipedia users and since Wikipedia runs mediawiki the support for Mediawiki would never ever die for as long as Wikipedia will be running (highly doubt they will migrate away). It lacked two things, propper LDAP support and access controls.
Dokuwiki was nice but nothing special, loved the fact that its file based.
When it came to presenting it to the rest of the company I had two options, choose Confluence or Mediawiki, I did not choose Twiki simply because it would have required too much time learning and customizing to please everyone. So I choose Confluence for its ease of use, feature sets and support and I choose Mediawiki for its familiarity to most (people hardly knew what a wiki was unless I added "its just like Wikipedia") and it was open source and free.
Convincing people that we needed a documentation system that was not built on word documents, text files, excel sheets and basicly whatever each person liked personaly was a pain in the ass. To get anyone else pumped up was impossible so I just stopped trying.
Instead I opted of installing Mediawiki (free) and just started using it myself. Slowly I managed to convince the few people in my department (network/system administration) to use it. I was still the biggest contributor but it became normal that the wiki was the first place to look when we had to search for documentation. After I added a what you see what you get editor it become even better to use.
Now, today, this year, I have switched to Confluence since they offer a 10 user package for $10, perfect to replace the Mediawiki setup. They even have a "universal" wiki converter to convert the old Mediawiki installation into Confluence, and I say "universal" since its just one way, from all major wikis into Confluence. I also added Gliffy for $10 for 10 users, no need for Visio anymore
The support for Confluence has been amazing, just go to http://support.atlassian.com/ and make a ticket and within 24 hours for even lowest priority tickets someone is there to help you. Best support by far I have received.
Confluence documentation is excellent and very detailed and about open source, Confluence is in a way open source, when you buy it you get the sourcecode as well which you are free to modify as you wish but you are not free to release it.
We also bought a 10 pack of Jira and the integration between Confluence and Jira is very nice.
The next step is convincing the rest of they company that they need this. Should be easy now after I have prooved that a wiki is a very powerful tool for documenting
To end this I just want to point out that when selecting a software solution you must take into account every aspect of having it around. How much it costs to start with, how much time you have to spend getting it to run, how it is to use it from the user perspective, how it is to use it form an administration perspective and if all fails, where do you get help. Open source is good in general, but for companies running open source software can be more costly than other solutions in the long run.
Cost is generally not the biggest issue. Your boss is probably against FOSS because most Pay-For-Play software generally comes with support & maintenance contracts issued from the people who wrote the software, which are extremely important to management types, while software like Plone requires a support or maintenance contract through a third-party provider (i found this: http://plone.net/providers )
If you can convince him that the best way for him to handle this situation and all potential future ones is purchasing a third-party support contract which can also be supported by you if need be since the software is open source, then you might have a shot. Otherwise, I'm not sure. I've seen a lot of good software packages turned down as solutions to business problems simply because there was no support contract.
Probability of an easy upgrade path if the product is discontinued.
# cat
Damn, my RAM is full of llamas.
Confluence is easy and looks pretty, but face it - the permissions controls are _abysmal_.
Sounds very much like a free PR ploy!! The differences are obvious!
"My employer is currently looking at adopting a content management system ... The candidates are currently Plone (OSS) and Confluence (proprietary, closed-source). For those with experience in each, what arguments in favor of Plone could be made to managers more interested in pragmatism than idealism?"
You hit the important point in your post: you cannot use idealism ("FOSS is neat!") to get your manager on your side.
When pitching any new idea to a manager, it's important to note the differences in priorities at each level. For examaple: with a line-level technologist (i.e. most geeks) technical knowledge is important, relationship-building and strategic planning are [generally] very unimportant. I'm talking about what's important to your job, not what's important to you.
But it's pretty much the opposite for managers, especially the further up the management chain you look. At the CIO level, strategic planning is the #1 most important thing, followed closely by interpersonal skills (CIOs often need to navigate lots of committees and meetings to get things done.) Individual technical skill is very unimportant to their work (some CIOs maintain some technical knowledge, but it's not used in their day-to-day work.)
I gave a talk about this very topic at Penguicon 2009, called "Linux in the Enterprise". My 1-hour talk used only a few conceptual slides, but you will be interested in the chart on slide #4. It presents this idea about "framing" very clearly. You can find the presentation at my blog, but the actual presentation is from May 2009. I think the data in the chart originally came from a Harvard Business Review study.
Your question indicated you were going to present this idea to a manager - I'm assuming (based on your wording) that you don't mean an IT Director or a Chief Information Officer, but instead an IT manager (probably your group manager.) Look closely as the chart from my presentation, and note the "Mgr" area of the graph. The "importance" lines all converge for the manager. That's not an error in the graph - that's indicative of managers stepping out of the "team lead" role, before they can make a transition to a "Director" role.
In the talk, I discussed how IT managers often have a hard time leaving their "tech person" role, moving into a manager role, because the IT skills that got them to the "manager" position won't carry them to the next level. And everything tends to have equal importance, because managers act as the filter between the technology teams and the director. So technical knowledge has about the same importance to their day-to-day job as budgeting and strategy. But I digress...
You'll need to make a case to your manager that addresses the points that he or she will find important, while at the same time thinking a step ahead to the level above your manager. Emphasize points that will be important to your manager's director. That means you need to de-emphasize the technology ("it's FOSS") and address the strategy. Does Plone fit into your organization's strategic IT plan? (Does your group even have an IT plan?) How easy can Plone integrate with other parts of the IT infrastructure? Can you tie Plone into your central authentication system? Who supports either package (patches, updates, new features, etc.) Your manager and director will not want to take on this effort, so be sure to mention commercial support options for Plone (usually this means "help desk"), the Plone developer community, and if it's available as a standard package in the Linux distro your sysadmins already run (meaning it would get updates by default as they regularly update the OS - which would be either good or bad, depending on your manager's preferences for testing after patches.)
It's not enough to say "FOSS is free". Especially because it really isn't. Talk about
For telling the truth...Confluence is open source.
Your real question is:
Convincing Your Employer To Go With Plone?
The answer to this depends on how good your organization is with Zope/Python. If you have onsite developers with Zope knowledge (who can support Plone), Plone is a no-brainer. And if you have developers familiar with other OOS software like Java, you have plenty of other products to choose from:
http://en.wikipedia.org/wiki/Category:Open_source_content_management_systems
http://www.cmsmatrix.org/
If you don't have any onsite development staff, the value proposition of OSS/Plone goes down because you will presumably have to hire someone to run it.
Frankly, that's what I would stress. If this is a large enough project you're going to have to hire someone to run it anyway. You can save on software costs by hiring someone who knows Plone.
If you're not hiring new staff it boils down to who within your organization is running the CMS and what THEY want. Most other considerations are relatively trivial. The more "out of the box" they need the software to be, the more that leans towards a proprietary solution. They might also want to be able to have a vendor to complain to and to provide direct support, again, proprietary has an edge here.
Popularity also factors in. I don't really know how popular Plone is, but Confluence is really popular. That means there will be lots of online resources (forums, FAQs, etc.) for Confluence that you might not find for Plone.
If you're willing to be pushy about it and it really matters to you, take advantage of the accessibility of F/OSS and learn about Plone setup and configuration over a couple weekends on your own. Go to your boss and tell him that if he's dead set on some other solution, you can do it, but there's no good reason not to use Plone; you're already competent with managing it, you know the features it has, you've set it up before, and if he okays it you can have the system deployed at no cost out of his pocket in a short period of time.
Again, this is a pushy way of handling things. Some bosses might not like it; some might see it as taking initiative. It would also feel like a pretty crappy investment of your time if it didn't work out, or you didn't learn the system as well as you claimed to have done. But I think the "I can already do it (for free) and I know it's good enough" is the strongest pragmatic argument you can possibly make.
Perhaps nothing else matters to you in your world, but I live in a world that includes my fellow English speakers who have a say in what English words and phrases mean.
With any other phrase other than "open source", you might convince me that the commonly held meaning of a phrase is actually wrong, and it originally meant something else, or in certain circles it means something else, or the Trobriand Islanders use it to mean something else.
But the phrase "open source" was invented to have a very specific meaning --go check the OSI website if you need to (and turn in your geek card)-- so you're not going to be able to say, "Well, but for me it means something else so I don't care if you end up misunderstanding because it's your fault!" So, what, someone else can come in and say "By 'open source' I mean that the room door is open as I look at the source code"?
If the source is open to you, but not open to anyone else, then you cripple the ability to locate and identify bugs. If the source is visible to people but not modifiable, you lose the ability improve the software as a community. If the source code is mostly but not completely visible to you, but you are not able to compile a binary that matches exactly the one that you are deploying, then you lose the ability to verify that the source code represents the software you are running. There is a reason for very specific requirements given to meet the criteria of "Open Source".
If you mean something else, go invent your own phrase.
404555974007725459910684486621289147856453481154 in hex is "You sank my Battleship?"
[GPG key in journal]
it's so flexible that you can do whatever you want with it :)
Say "don't listen to me, I'm too biased in favor of open source software that I'm not really doing a legitimate comparison. In fact, maybe you should fire me."
Seriously, though, you need to research the two products, look seriously into support, total cost (not just initial sticker), compare functionality, and do all the other due diligence that you should be doing.
On the Plone side, for some of that you probably need to be talking to Enfold Systems or one of the other companies providing professional paid support and so forth for Plone.
(I'm assuming that things like support contracts are important to your bosses. Maybe they're not. I know they would be for mine.)
Asking Slashdot probably isn't the right answer. This isn't fun and games philosophical shopping time, boys. This is work and business.
I'll just sit waiting to be modded "asshole" or told how wrong I am in some fashion by other open source fan boys.
The preferred solution is to not have a problem.
... they should have renamed openoffice to xmloffice. It would have finessed Microsoft's "Office Open XML" confusion strategy, AND been Buzzword 2.0 compliant.
Wikipedia is usually one of the first results for any search. Show them how good the wiki search is, how easy it is to create new pages, links, headings, lists, etc, and how quickly they can find anything through the search box.
I don't know if they use it anymore but when I was involved about 4+ years ago they were converting their whole intranet to Plone. The system they developed to the various community groups to coordinate and work together was pretty remarkable. If you do a search for Burningman and Plone on Google you'll find some resources.
Also there is a guy in the Plone community who goes by the handle "Spanky". Reach out to him, he's a good guy and may be a good resource.
that's a good exercise for a techie: you now have to sell the OSS solution you like to your management. A couple of pointers:
1- Are you sure it *is* a good solution to the issue at hand ? why ?
2- What do you think Management will like about it ?
3- What do you think Management will *not* like about it ?
4- What are the advantages of your solution management is overlooking ?
5- What pedagogical approach can you adopt to not try and force your solution down management's throat, but make them ask for it ?
Hint: most of your arguments must *not* be technical in nature, probably not even functionnal, but financial and political. Talk references, money, interoperability, uptime...
The Cloud - because you don't care if your apps and data are up in the air.
When it's a business, it's whatever works best. Choosing idealism over pragmatism is not something you should do when other people's jobs are affected (unless your ideals include "others are obligated to take the risks my personal ideals dictate"). If the closed-source solution is objectively better, you use it. If the open-source solution is objectively better, you use it. If it's objectively a tie (which should almost never actually occur, but...), then AND ONLY THEN should idealism be the deciding factor, and even then the question becomes "whose ideals". You're apparently not the boss, so why should your ideals trump those of your boss or the business owner?
Customer contact management is important. It's no place to be messing around. There is no problem with FOSS that is the best in its category, or if total cost of ownership including licenses/support is too high for commercial software. However, FOSS doesn't get a free pass if it isn't rock-solid, usable by the people who have to live with it, is interoperable with other software, and supportable.
Make your evaluation based upon facts. If you can't make a fact-based, backed-by-the-numbers argument in favor of Plone, you shouldn't be advocating it.
And vice versa. The horror stories I could tell about Jira/Confluence...
Everybody gets what the majority deserves.
They all have annoying quirks because they are an abstraction of what is going to appear on the page. That's the price you pay for extra convenience of not having to do full desktop publishing.
A spreadsheet is the wrong tool when you have 65535 rows.
The cost of MS Office are relatively small... For a mid-size or larger company the cost of an MS Office license is about $120 or so (depends on volume and license specifics, with the purchase of a new compnter it's nearly free). Assuming that a company rebuys a license every 2 to 3 years (for upgrades) the cost per year is about $50/employee per year. Assuming that the employee is paid $35K per year, Office cost about 0.14% of their wages. (A lower percentage for higher paid employees of course). You don't need to show much of a productivity improvement to justify the MS Office investment. A day of "conversion" training will cost $175 (35,000/200 working days). That pretty much blows the cost savings.
The office chair that the employee is sitting in costs more than 2 copies of Office, yet I hear few OSS enthusiasts arguing for Open Source Office Chairs (OSOC). Yep, your company could probably save money by giving new employees the OSS equivalent of an office chair--some cinder blocks and a piece of board. But would it really save money?
Although I'm a fan of much Open Source software, I just can't make the economics of replacing MS Office work.
Is there money in the current budget to buy a content management system?
Even if there is, the way to pitch open source is to go ahead and install a prototype system on a spare computer. Call it an "extended evaluation", or whatever. Set up Plone as you would if you were going to use it for the actual project. If it works well, then it becomes the winner by default. Since this is not "trialware", you don't have that "30-day" BS to worry about.
The pitch to management is this: "If we can save money by doing this, then let's do it. If by some chance the effort fails, we can always fall back on the more expensive option. The worst possible outcome is that we make mistakes that we learn to avoid during the real implementation."
This puts open source products on the fast track to adoption, since you don't need budgets or expenditures to get the ball rolling. Managed properly, IT gets credit for getting things done without a perpetual beg-a-thon for more money.
Go-oo.org is a fork managed by some folks at Novell that incorporates multiple patches that haven't made it into the main branch yet. One of those is exceeding 65K rows (now 1048576). It's not in the main branch because there are apparently some problems with calculation performance with many rows and some problems with positioning with drawing objects. More details in http://kohei.us/2010/02/20/increasing-calcs-row-limit-to-1-million/
fencepost
just a little off
Linux...IBM uses this.
OpenOffice...Oracle was involved in developing this.
So on and so forth.
PHBs won't understand reasoned arguments about the benefits of F/OSS versus proprietary software. They will understand that a successful Fortune 500 company trusts their stockholders' profits to it.
LK
"Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
The question you need to ask is, "Is the open source solution better?"
If it is, explain the reasons why to your boss.
If it's not, then you should use the other solution.
"Being open source" is irrelevant for end users. They want to use the software, not fuck around with its source code.
Confluence is better. Sorry about that, OSS.
For support documentation, why not just use Mediawiki? We've got a team of about 8 in IT support and infrastructure here. We're tried using Zope and Plone in the past and they were utter failures. Mediawiki has worked fantiastically. I think the main issue there was mostly that it just took too much effort to update or create docs in Zope. It's very hard to get people to write documentation; it's very hard to get them to correct, update and maintain it. You want as few barriers in there as possible. To be honest, I'm not sure FOSS really comes into the equation here.
At my last company I did an informal one-on-one query of my coworkers regarding why they didn't put stuff into Confluence, and they echoed precisely these sentiments, point-for-point. I found the XML::RPC-based API pretty good, but was thinking it would have worked much better as a central information repository if paired with a Google search appliance -- assuming that works as well as one would think.
We were looking to move a few dozen sites off of a proprietary Web CMS and one of the reasons we decided on popular OSS solutions like Wordpress, Drupal, Joomla, etc. is that it is pretty easy to find vendors who know them. We lease virtualized servers at Rackspace so we control the hosting, and then we hire on vendors to help us build and support our sites.
My metaphor was that going with a hosted, proprietary CMS was like leasing a car where you had to always pay the dealer for work. Hosting our own Wordpress or Drupal site is like owning a car--if we don't like our mechanic we can get a different one.
It has already paid off a couple times--we soured on a vendor and were able to easily find and bring in a new one to support the site.
One reason we decided against Plone is that it is not heavily used or supported in my area (DC).
Build a man a fire, he's warm for one night. Set him on fire, and he's warm for the rest of his life.
Tell your company to seek the source code of the closed source CMS.
I'd like to buy homeland for our 10 million people. http://twitter.com/mahadiga
My dentist is always trying to get me to floss. I find Glide works best.
Point out that the vendor can and will kill off a product and support for that product OR charge like a wounded bull for specialised support OR that the company may fold, and that they are not legally obligated to continue a product that the company may become dependent up. Then point out that in the case of open source, it is possible to hire someone to develop the product further and support it, and that even if there is a cost penalty it won't be extortionate.
All other arguments are a waste of time for mission critical applications. Open source may or may not be cheaper.
These posts express my own personal views, not those of my employer
I second that. One should also be careful that people don't just use it as a farm for word documents. That defeats the whole purpose of a Wiki.
Being FOSS is a strong reason in itself to prefer that software. Especially when the whole company plans to depend on it for internal communication. Relying on proprietary software means giving up control on your business.
Funny how it's fine to push an MS agenda. You know, all that "Nobody got fired for buying Microsoft" or "license costs are trivial" or "you're biased against CSS" (which is required when your management are biased FOR it). But several people have spouted "stop with the bias first".
Funny that.
We've switched about half our users to 2007/2010 (mostly with the purchase of a new machine), and after initial hesitation most users became quite positive about the new interface. After 2 weeks or so, they all reported it was easier to use and commands were more obvious. We do minor customization of the quick access tool bar to put some very common-to-us commands up there.
However, I have to say that 2010 is much better than 2007 (it just seems easier to use). I also wish they had added more functions to excel in addition to more colors, rows and columns. Excel really needs a built-in regex (not via macro) and a find last function.
What about the obverse? Does the closed source product tick all those boxes? After all, one *fake* one is MS Word compatability and how when you get a docx from a customer (or give one to them) it won't look right. Except this is true with MS Word too. IT doesn't manage to be 100% compatible.
Then you have the "tickbox" "must work with $SOME_CLOSED_PRODUCT". Why?
And also "Your $PRODUCT doesn't have $FEATURE" when $FEATURE isn't wanted, needed or even usefully implemented (CMYK separation, for example).
So, does the closed product tick all those boxes? Does it tick them all when you've picked the boxes so discriminate against closed source (e.g. "Must be FOSS": get a MS product that ticks THAT box!).
Pragmatisim deals with facts, reality, idealism does not. To advocate for idealism over pragmatism in the workplace is to advocate for fantasy over reality - that may not be a good long-term career strategy. Let the products stand on their merits, warts and all, and let your employer make the best informed decision you can.
Ken
With open source just download and try how easy are it to use I didnt discovered CMS which gives better results in less time than writing pages in your favorite framework
From my experience month ago I discovered drupal as open source and little else.
I abbadoned idea after week with little actualy done. Most of time I spend reading really unhelpfull howtos, figuring why navigation item is on page twice, unsucessfully tried to change login form to one-line and numerous other issues.
In some forum I found description like this
Time to create pages from scratch:14 days
Time to create pages with drupal: 3 days (+ 3 months to learn drupal)
Comment removed based on user account deletion
You seem to be implying that, even prior to the invention of the term "open source", it already had a meaning, but this is not the case: the term was created at a meeting of the minds http://www.catb.org/~esr/open-source.html specifically so that we could avoid this sort of mixup and not be accused of "redefining the language".
I already addressed this in the post to which you are responding; when you reply to me with an answer I already anticipated, you're supposed to address that, too. I had given an example of someone defining "open source" as meaning that the room door is open as you look at the source code --are you going to turn around and say that that use of the word "open" is invalid where yours is valid? Please see http://web.archive.org/web/20060423094434/www.opensource.org/advocacy/faq.html Prior to that, the technical term was just used by spies to denote publicly available info, and was not even used in the software world.
But most telling of all is your apparent indifference to the way the software community is using the term "open source". When you say "nothing else matters", what you are saying is not just "the word 'open' already has a meaning" (ignoring its juxtaposition with the word "source") but also "I don't care how the rest of you use the English language as a mutually agreed-upon way of communication".
Don't agree? You'd need to cite a use of the term "open source" prior to February 1998 to mean what you say it means. Or else I'd ask you to get stuffed and take a hike. Well, of course I meant "obtain some stuffing and grab a fee increase"! What, are you saying that words have different meanings when used in certain specific combinations?
404555974007725459910684486621289147856453481154 in hex is "You sank my Battleship?"
[GPG key in journal]
Mediawiki ships with a user-friendly rich text editor. You only need to change one line in the config file to enable it. I installed a copy last week to use internally at work, and when someone asked, I enabled it.
2 weeks? Time is money. By your figures above that's $1750.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
You need a database, not Excel or OpenOffice....
StarOffice is the commercial product and has commercial support contract options.
I tried to sell Plone as an alternative to SharePoint once, everyone was very impressed with Plone and shocked at the hardware requirements of SharePoint (and shitty performance of the VirtualServer setup the Microsoft salesmen demoed!) and the fact that the whole company would have to upgrade from MSIE6.
The factor that killed Plone was no Single Sign On via AD, which SharePoint had out of the box of course.
There was a commercial Plone plugin for SSO which required Samba3 and all sorts of OpenLDAP hacks (or it had to run on Windows) to even partially work, but then the FOSS argument and reuse of existing *Sun* hardware went out of the window.
Until Plone integrates with Enterprisey vendor lock-in stuff like AD/Exchange, it isn't going to be gaining any traction. Still you've got to give it to Microsoft, SharePoint is rubbish but somehow they still manage to sell it.
#include <sig.h>