[NBLUG/talk] Linux in offices initiative

Omar Eljumaily omar at omnicode.com
Fri Aug 29 15:43:04 PDT 2014


Thanks everybody for some excellent commentary on this subject.  I will 
address individually some of the issues soon, but wanted to weigh in on 
a couple of things right now.

1. Industry wide, and not just for Linux, if IT pros want to start being 
taken seriously like doctors, lawyers, engineers, and even mechanics and 
construction contractors, there's going to have to be some move towards 
standardization, and Linux procedures, I think, is a good starting point 
for it.  Some IT work is arguably more complex than brain surgery, yet 
you can find help wanted ads on Craigslist literally offering $5/hour 
for IT work.  It's crazy, there's too much chaos in the field.

2.  The confusion also exists among customers, and I think there's a lot 
of missed opportunity there because of it.  There would be way more 
money in the industry if there weren't so much confusion. Whether or not 
it's a good fit for the Linux ecosystem, standardization and 
organization would benefit both vendors and consumers for Linux services.

3. Sorry, I'm still harping this cloud vs LAN server issue.  I think 
it's central to what people want to do with Linux, and whether or not 
there's going to be a such thing as a small office server in the 
future.  I look at the issue as the "App Model" vs. the "Machine 
Model."  Do companies utilize a collection of Apps from various vendors 
or do they utilize a comprehensive, organized machine?  I think if it's 
just apps, then there's no future for office servers. If it's a machine, 
where apps need to be organized and communicate with each other, it's 
the "Machine Model" and office servers make sense, and the only 
realistic way to do it is with Linux.

Thanks,

Omar


On 8/29/2014 2:04 PM, Jordan Erickson wrote:
> On 08/29/2014 01:29 PM, Steve S. wrote:
>> Agreed on all points, Jordan!
>>
>> I'll just add one more point, and expand one...
>>
>> You wrote, "My clients know that I am the one that
>> supports their network, and that is their bottom line."
>> I'd expand that to the notion that clients want an
>> array of options for support; not to say they aren't
>> happy with whatever provider they've got, but most
>> businesses like to know that there's always a backup;
>> other providers who can step into the breach if their
>> current provider goes out of business, larger providers
>> who can service out-of-area if they expand, etc etc etc.
>>
>> There's pretty-much always another shop with more
>> MS techs, anywhere they look.  Not so much for Linux,
>> so it can take a bit of extra "selling" to show them that
>> the downsides are more apparent than real, and the
>> upsides are very real indeed.
>>
>> This veers back toward the OP's point about NBLUG
>> discussing "best practices,"  If we all had the same
>> few core distro's we preferred, the same few solutions
>> to each of the substantive business-challenges (and
>> if the Linux world outside the NorthBay agreed with
>> our selections) then the linux choice would be much
>> easier for businesses to make...
> What you're talking about doing is standardizing on a "business Linux"
> distro of sorts, one that is uniform across all fronts. This is the
> exact opposite of what the Linux ecosystem actually is. That being said,
> I don't see that there already isn't "standard practices". What I see
> most people are used to thinking is, "This looks different, I don't know
> how to use it". But with Linux, what people normally see in a GUI tool
> is, many times, simply a graphical wrapper for a shell tool.
> Many-a-backup programs are simply based off of tar, rsync, etc. for
> example, and in my humble opinion, learning the underlying tools is what
> counts when you want to become fluent enough in Linux to be able to
> support production environments with it.
>
>
>> You also called out the MS "ecosystem," by which I think
>> you meant the self-sufficient all-MS-product office, without
>> need for any non-MS product running on their hardware &
>> working without putting effort into "compatibility" between
>> two different software packages.
> Not necessarily. Most businesses always have the weird, quirky apps that
> M$ didn't make. What I meant by the ecosystem is that the core of their
> computer network (and many of their business partners as well) is M$
> dependent. Need Internet Explorer 11 to log into the new version of your
> partners billing site? Well, you'll have to upgrade to Win7 for that.
> Can't open that .xlsx file that just got e-mailed to you? Upgrade
> Office. In the end, it always ends up that you have to pay Microsoft
> $$$. Also, the contrast between how single-purpose utilities are
> drastically different when you're talking about M$ vs. Linux ecosystems.
> For example, your Outlook .PST file got corrupted? There are LOTS of
> options out there, and 90% of them cost at least a couple hundred bucks
> from third party vendors. It makes sense - if you've paid a few hundred
> for Office, and another couple hundred for Windows, of course you'll
> shell out some more to get it working correctly again.
>
> That's what I meant by the ecosystem.
>
>
>> The other issue I see is a b2b-level compatibility.  Most
>> businesses want 100% compatibility with whatever their
>> customers/clients/vendors/regulators/partners/etc use.
>> When they send out a document, they need to know that
>> the recipient will see exactly what they saw; when they
>> open a document from elsewhere, they want to be sure
>> they're seeing it exactly as-written.  In my wife's office, for
>> example, there's a problem even with different versions of
>> MS-Word (MS-Win vs OS X), which occasionally display
>> somewhat differently.  OOo (which they've seen & used)
>> is a complete non-starter because it is so very different in
>> its output.
> I see this as a diminishing issue, at least with document compatibility.
> I haven't been in too many situations where people are e-mailing Word or
> Excel docs back and forth to people outside their own office for
> editing. The norm that I've seen is PDF export, which seems to work very
> well, including from OOo. Go figure, Adobe! ;)
>
> Going back to an earlier point, I actually think that standardizing on a
> single Linux distro, and congregating around a single set of tools to
> accomplish standard business tasks could actually end up hurting what
> makes Linux so strong in the first place. I've been around long enough
> to see that as soon as you take away the element of choice, you
> ostracize the community that would otherwise be helping make it better.
> I've seen this with numerous Ubuntu/Canonical initiatives. Not to say
> that it isn't a good distro, but there have been a lot of hard feelings
> generated when Shuttleworth has layed his iron fist down and told the
> community that 'this is the way it's going to be'. He has to remember
> that diversity is the key to the success of Linux, as does everyone I
> think. When you take that diversity away, problems arise.
> _______________________________________________
> talk mailing list
> talk at nblug.org
> http://nblug.org/cgi-bin/mailman/listinfo/talk



More information about the talk mailing list