[Osmf-talk] Framework for the foundation's hiring practices

Allan Mustard allan at mustard.net
Tue May 12 02:55:52 UTC 2020


Emilie,

Do you consider maintaining over 90 servers in a global network with
only two part-time volunteer sysadmins to be low-risk and sustainable in
the longer run?

When I ran the computer center for the Foreign Agricultural Service 20
years ago, a small U.S. government agency with 1,350 seats and a few
dozen servers, we had a dozen fulltime network techs, all of them
accredited as sysadmins, so that we would have plenty of redundancy to
handle travel, vacations, sickness.  They put in lots of overtime
keeping our global network up and running.  I don't know about you, but
OSM's lack of redundancy makes me nervous. 

You wrote:

> I am not a big fan of hiring people without properly thinking it through
We are doing exactly that, thinking it through, and soliciting ideas
from the OSMF membership.  Your suggestions, please?  The Board wants
your ideas on how best to deal with the shortage of volunteers for
critical functions.

apm

On 5/11/2020 5:55 PM, Emilie Laffray wrote:
> Hello Allan,
>
> I will mostly say two things, one short, one slightly longer. 
> First, my name is not Emile but Emilie. If you have to butcher it,
> please use Emily. 
> Second, SMART and agile are not really on the same level. I want
> nothing to do with SMART in a system like the OSM foundation to be
> honest. I think it is a waste of people's time in the first place.
> Also, I am quite incensed by the juxtaposition of opposing SMART to
> opposing the need of well defined goals in the first place. 
> I kinda disagree with operations not being able to work in an agile
> environment (I use Kanban at the office for exactly that reason) but
> that would be besides the point.
>
> In the end, I agree with a lot of points that Frederick raised and I
> am not a big fan of hiring people without properly thinking it
> through. I strongly believe in paying people to do some jobs like
> accounting and so on (and even pushed for it in the past). I am not
> sure we want to do so in operations without some kind of supervision.
> And supervision takes time.
>
> Anyway,
> Emilie Laffray
>
>
>
>  
>
> On Mon, May 11, 2020 at 5:40 PM Craig Allan <allan at iafrica.com
> <mailto:allan at iafrica.com>> wrote:
>
>     Emile: "oh hell no" is a fair expression of horror!  The buzzword
>     is hateful and often mis-applied - but do you hate the general
>     idea of pinning routine work and performance assessment down to
>     pre-agreed work goals and times? If you're spending other people's
>     money, I hope not.
>
>     I fully agree that there are other ways to work. Agile development
>     is far looser, allows frequent direction changes and is just far
>     more fun. In this case we are not hiring people to do development,
>     we are working to reduce risk and improve the resilience of our
>     operations.  In my world, Operations are not agile - for good
>     reason.  The proposed Ops contractor is there to follow the
>     Standard Operational Procedures to the letter, and keep the
>     engines running at 99.999% availability. 
>
>     In this way our dedicated  SysOps and SysAdmin volunteers are
>     released from the daily drudgery of monitoring and maintenance
>     work and can take on the more interesting systems development work.
>     CA
>
>
>     On 2020/05/11 22:42, Emilie Laffray wrote:
>>     Just a quick "oh hell no" to SMART. It is one of the worth
>>     assessment methodology in the tech world.
>>     This is the bane of my life and you end up creating goals just
>>     for the sake of creating goals that are impossible to miss. 
>>
>>     On Mon, May 11, 2020, 14:15 Craig Allan <allan at iafrica.com
>>     <mailto:allan at iafrica.com>> wrote:
>>
>>         Hi All,
>>         I was battling to find a short reply to Tobias's request for
>>         comments,
>>         but Severin has crystallised my thoughts.
>>         We are being asked to comment in the absence of specifics
>>         which IMHO is
>>         not a good approach - I predict it will produce lots of words
>>         and very
>>         few answers. So here are my words:
>>
>>         The situation seems to be serious. We are running a world
>>         wide network
>>         of about 96 servers, using only 4 volunteers as sysadmins.
>>         The system
>>         is, as far as I know, very heavily used and is only just
>>         coping with the
>>         loads. So I pause to salute the heroes who keep it all
>>         running. Amazing
>>         work!
>>         I fully support buying some permanent resource to add
>>         resilience to the
>>         system and take the load off the volunteers.
>>
>>         I also fully support a lot of the concerns people have posted
>>         about the
>>         messiness of directly hiring somebody who is going to work for a
>>         collective, not work for one manager.  To eliminate a lot of the
>>         problems I would support arms-length hiring practice. By that
>>         I mean we
>>         should be hiring a medium to large company to perform the
>>         services.
>>         Hiring a company is typically more expensive at first
>>         impression, but
>>         the extra we will pay to a company rolls up and packages all
>>         the legal,
>>         labour and human messiness and hidden costs of employing an
>>         actual person.
>>
>>         Hiring a company has the added advantage that it forces us to
>>         write a
>>         clear SMART job description, forces us to limit the scope of
>>         work and it
>>         encourages us to do an unemotional assessment of performance
>>         against
>>         that clear job description.
>>
>>         I also think that a company performing a set task will be
>>         focused on the
>>         work, will be less involved in the totality of the OSMF and
>>         will have
>>         less inclination to mess with our internal politics.
>>
>>         Craig ALLAN
>>
>>         * My mapperId is cRaIgalLAn
>>         * SMART is an acronym for "Specific, Measurable,Attainable,
>>         Relevant,
>>         Time-bound"
>>
>>
>>
>>         On 2020/05/06 22:24, Tobias Knerr wrote:
>>         > Hi all,
>>         >
>>         > the OSMF Board wants to think about a general framework to
>>         hire people
>>         > to fill in the gaps that volunteers can't fill. We believe
>>         that, given
>>         > good practices and firm boundaries, hiring people would be
>>         worthwhile.
>>         > It could ensure the continued stability of the OSM platform
>>         (servers,
>>         > integral software) among other things, and augment the
>>         currently
>>         > overworked volunteers and under resourced efforts in the
>>         face of
>>         > continued growth.
>>         > ...
>>         >
>>         > Feel free to share your ideas here or send them to
>>         board at osmfoundation.org <mailto:board at osmfoundation.org>
>>         >
>>         >
>>
>>
>>         _______________________________________________
>>         osmf-talk mailing list
>>         osmf-talk at openstreetmap.org <mailto:osmf-talk at openstreetmap.org>
>>         https://lists.openstreetmap.org/listinfo/osmf-talk
>>
>
>     _______________________________________________
>     osmf-talk mailing list
>     osmf-talk at openstreetmap.org <mailto:osmf-talk at openstreetmap.org>
>     https://lists.openstreetmap.org/listinfo/osmf-talk
>
>
> _______________________________________________
> osmf-talk mailing list
> osmf-talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/osmf-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20200511/c289d7af/attachment.htm>


More information about the osmf-talk mailing list