PM Activities for Firefox 3

classic Classic list List threaded Threaded
28 messages Options
12
Reply | Threaded
Open this post in threaded view
|

PM Activities for Firefox 3

Sherman Dickman
All,

In case I haven't met some of you, my name is Sherman Dickman, and  
I'll be driving product management activities in support of Firefox 3  
planning.

Since the product management role is a relatively new one at Mozilla,  
I thought it would be useful to outline some of the most important  
functions that a PM should perform, particularly within the context  
of product planning.  Many of these will be conducted in parallel  
with the Gecko 1.9 engineering planning efforts currently underway.  
The end PM deliverable for Firefox 3 will be a MRD that can be  
referenced by all, but the real value for the Mozilla community will  
be derived from the MRD development process itself.

Activities
------------------------------------------------------------
Here are some of the activities and deliverables that a PM helps to  
coordinate, all geared towards providing the organization with the  
information it needs to make effective product decisions.  These  
include:
      * identifying market and technology trends
      * competitive analysis and strategy
      * customer segmentation and value delivery chains
      * user personas and use case scenarios
      * value proposition and differentiation strategy
      * product vision and strategy
      * key feature requirements
      * participation in release criteria discussions
      * etc.


Subjective Activities
------------------------------------------------------------
Additionally, there are more subjective activities that can help  
clarify thinking and spur creativity.  One of the most important  
things a PM can do is to engage the team in answering some very tough  
questions about our goals, product, and strategy. For example:

* Where do we want to go?  How will we get there?  Why do we think  
we'll be successful?
* Why should anyone use our product over another product?  Do the  
benefits outweigh the tradeoffs that a user must endure when  
switching?  If yes, why don't more people use our product?
* Which of our technology assets are clearly superior to the  
competition?  How do we keep them from being easily copied or  
duplicated?  Are there barriers to entry or significant hurdles for  
competitors to overcome?

Questions such as these can be particularly difficult to answer.  But  
once answered, they help to direct downstream activities tremendously.


User Feedback
------------------------------------------------------------
Another critical product management function is to collect user data,  
and then analyze the information to find hidden stories and  
opportunities.  The biggest challenge in collecting user data is  
knowing what questions to ask and from whom.

If you ask a panel of home cooks questions about how to build a  
better oven, they'll tell you. But if you ask them about the  
challenges of feeding a family each night, you might receive a  
completely different set of answers — answers that might suggest a  
new product opportunity (such as the microwave oven).

A formula that has worked very well in the past looks like this:

    1. How do we build a better product?
    2. What are your biggest challenges?  What do you spend most of  
your time and/or money on?  What keeps you up a night?  What would  
you do differently if given unlimited resources?  What's your worst  
case scenario?
    3. If we did everything that you told us to do in question 1,  
would that address any of the issues outlined in question 2?

The answer to the last question is usually "no," but that's okay if  
the customer articulates their greatest pain points and unmet needs,  
many of which *can* be addressed by the product if approached in the  
right way.  For example, Nike golf shirts now have SPF 30  
protection.  What types of questions did they ask consumers to come  
up with that idea?

The product management team at Mozilla is currently working on  
mechanisms that will allow us to better capture user data to uncover  
these hidden stores and opportunities.  Expect to see some things  
roll out in the next month or so.

Cheers,
Sherman

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

pascal chevrel
Sherman Dickman a écrit :

> All,
>
> In case I haven't met some of you, my name is Sherman Dickman, and I'll
> be driving product management activities in support of Firefox 3 planning.
>
> Since the product management role is a relatively new one at Mozilla, I
> thought it would be useful to outline some of the most important
> functions that a PM should perform, particularly within the context of
> product planning.  Many of these will be conducted in parallel with the
> Gecko 1.9 engineering planning efforts currently underway.  The end PM
> deliverable for Firefox 3 will be a MRD that can be referenced by all,
> but the real value for the Mozilla community will be derived from the
> MRD development process itself.
>
> Activities
> ------------------------------------------------------------
> Here are some of the activities and deliverables that a PM helps to
> coordinate, all geared towards providing the organization with the
> information it needs to make effective product decisions.  These include:
>      * identifying market and technology trends
>      * competitive analysis and strategy
>      * customer segmentation and value delivery chains

I will be very interested in seeing how you will draw a diagram for the
mozilla value delivery chain, given our mostly community-based and
informal production model, it will be like modelizing social networking,
good luck :)

>      * user personas and use case scenarios
>      * value proposition and differentiation strategy
>      * product vision and strategy
>      * key feature requirements
>      * participation in release criteria discussions
>      * etc.
>
>
> Subjective Activities
> ------------------------------------------------------------
> Additionally, there are more subjective activities that can help clarify
> thinking and spur creativity.  One of the most important things a PM can
> do is to engage the team in answering some very tough questions about
> our goals, product, and strategy. For example:
>
> * Where do we want to go?  How will we get there?  Why do we think we'll
> be successful?
> * Why should anyone use our product over another product?  Do the
> benefits outweigh the tradeoffs that a user must endure when switching?  
> If yes, why don't more people use our product?

That's an easy one : because they have never heard of it. Even in Europe
where we have a significantly higher (and growing) market share than in
the US, firefox users are mostly very well informed people with excelent
technology skills, that's much wider than the geek community but that's
also very far from the general public and probably won't go further than
about 20% of people with an internet access. People can make a choice
only if they know that this choice exists.

> * Which of our technology assets are clearly superior to the
> competition?  How do we keep them from being easily copied or
> duplicated?  Are there barriers to entry or significant hurdles for
> competitors to overcome?

I think I don't agree on this strategy.

> User Feedback
> ------------------------------------------------------------
> Another critical product management function is to collect user data,
> and then analyze the information to find hidden stories and
> opportunities.  The biggest challenge in collecting user data is knowing
> what questions to ask and from whom.
>

>
> A formula that has worked very well in the past looks like this:
>
>    1. How do we build a better product?
>    2. What are your biggest challenges?  What do you spend most of your
> time and/or money on?  What keeps you up a night?  What would you do
> differently if given unlimited resources?  What's your worst case scenario?
>    3. If we did everything that you told us to do in question 1, would
> that address any of the issues outlined in question 2?
>
> The answer to the last question is usually "no," but that's okay if the
> customer articulates their greatest pain points and unmet needs, many of
> which *can* be addressed by the product if approached in the right way.  
> For example, Nike golf shirts now have SPF 30 protection.  What types of
> questions did they ask consumers to come up with that idea?

"how much would you accept to pay for this upper-class recognition
symbol ?" I guess ;)

To be honest I don't think this method can be directly translated to
Firefox because :

- we don't have customers but users, that's a huge difference in the way
you communicate and interact with the people. We are offering something
much closer to a state service like the NHS or public transportations
than a targeted commercial product.
- asking what users want is ok when you address an identified and
well-known niche market like the above Nike example, not when you build
a mass-market service.


>
> The product management team at Mozilla is currently working on
> mechanisms that will allow us to better capture user data to uncover
> these hidden stores and opportunities.  Expect to see some things roll
> out in the next month or so.

I can only ask you one thing : be very open about it. I and many others
still remember the absurd UI and feature choices Netscape/AOL did, we
thought them in Bugzilla and were usually said 'our secret internal
usability study says that the users expects this'. The Phenix project
contradicted all AOL internal marketing studies, user personas and use
cases...

Ask a non firefox user what he expects from a browser and he will
describe his conception of a browser based on his knowledge of his own
current solution : IE. You will never hear from joe user that he needs
tabs, XUL or extensions, should we remove them then ? You will hear from
him that we need the googlebar, access to activeX based sites like
windows update and full hotmail/yahoo mail compatibility, should we
modify the product for that compatibility then ?

Here are a few very obvious things that don't need a study to be
discovered and should be top priority in my opinion :

- we don't have any market share in arabic countries because there is no
official arabic mozilla portal nor any arabic online resources about
mozilla.
- one of the next 2.0 release main features (which is a problem for 3.0
as well), form spellchecking, will only be available to English users
for licensing reasons. It's a 6 year old problem. There is an urgent
need to finance the creation of our own dictionnaries with a licence we
and openoffice can use, because they have the same problem.
- we need corporate deployment and *maintenance* tools to get into the
enterprise. The CCK is a first small step, but the one major blocker for
network administrators is maintenance. I have to say that I am quite
concerned to see that all you said seems to be only based on the
end-user and nothing is said about strategies for the corporate world.
- plugins are still a problem, it has improved a lot over the Suite
days, but we definitely need to make plugins installation easier.
Something like what they did with EasyUbuntu would be great I think.
- native ODF compatibility would be a corporate killer feature for FF3

Cheers

Pascal
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

dolphinling-2
In reply to this post by Sherman Dickman
Some of my thoughts while reading this:

Sherman Dickman wrote:

> All,
>
> In case I haven't met some of you, my name is Sherman Dickman, and I'll
> be driving product management activities in support of Firefox 3 planning.
>
> Since the product management role is a relatively new one at Mozilla, I
> thought it would be useful to outline some of the most important
> functions that a PM should perform, particularly within the context of
> product planning.  Many of these will be conducted in parallel with the
> Gecko 1.9 engineering planning efforts currently underway.  The end PM
> deliverable for Firefox 3 will be a MRD that can be referenced by all,
> but the real value for the Mozilla community will be derived from the
> MRD development process itself.
>
> Activities
> ------------------------------------------------------------
> Here are some of the activities and deliverables that a PM helps to
> coordinate, all geared towards providing the organization with the
> information it needs to make effective product decisions.  These include:
>      * identifying market and technology trends

And also how they relate to us, and whether we can be helpful by doing
something about them.

>      * competitive analysis and strategy

In what way? We're trying to be as good as possible here, not "better
than the competition".

>      * customer segmentation and value delivery chains
>      * user personas and use case scenarios
>      * value proposition and differentiation strategy
>      * product vision and strategy
>      * key feature requirements
>      * participation in release criteria discussions
>      * etc.
>
>
> Subjective Activities
> ------------------------------------------------------------
> Additionally, there are more subjective activities that can help clarify
> thinking and spur creativity.  One of the most important things a PM can
> do is to engage the team in answering some very tough questions about
> our goals, product, and strategy. For example:
>
> * Where do we want to go?  How will we get there?  Why do we think we'll
> be successful?

How about: Where do we want to go? Why do we want to go there? What's
the best way of getting there?

In my mind, when you're trying to help people being "successful" is
never possible, only "as good as possible".

> * Why should anyone use our product over another product?  Do the
> benefits outweigh the tradeoffs that a user must endure when switching?  
> If yes, why don't more people use our product?

Yes, but in context. If someone would benefit from switching but hasn't,
we should identify why they haven't (don't know about it? fear?
misinformation?) and promote it in a way to overcome that. If they
wouldn't benefit from switching, we should identify why not (switching
is too hard? missing critical feature?) and work to fix that.

> * Which of our technology assets are clearly superior to the
> competition?  How do we keep them from being easily copied or
> duplicated?  Are there barriers to entry or significant hurdles for
> competitors to overcome?

I agree with pascal here, but will be more vehement: Absolutely not.
Mozilla's mission is to promote choice and *innovation* on the internet.
Our features can't be improved on if they can't be copied, and with no
improvement there's no innovation.

> Questions such as these can be particularly difficult to answer.  But
> once answered, they help to direct downstream activities tremendously.
>
>
> User Feedback
> ------------------------------------------------------------
> Another critical product management function is to collect user data,
> and then analyze the information to find hidden stories and
> opportunities.  

Yes, but opportunities for users to be helped, not as opportunities for us.

> The biggest challenge in collecting user data is knowing
> what questions to ask and from whom.
>
> If you ask a panel of home cooks questions about how to build a better
> oven, they'll tell you. But if you ask them about the challenges of
> feeding a family each night, you might receive a completely different
> set of answers — answers that might suggest a new product opportunity
> (such as the microwave oven).
>
> A formula that has worked very well in the past looks like this:
>
>    1. How do we build a better product?
>    2. What are your biggest challenges?  What do you spend most of your
> time and/or money on?  What keeps you up a night?  What would you do
> differently if given unlimited resources?  What's your worst case scenario?
>    3. If we did everything that you told us to do in question 1, would
> that address any of the issues outlined in question 2?
>
> The answer to the last question is usually "no," but that's okay if the
> customer articulates their greatest pain points and unmet needs, many of
> which *can* be addressed by the product if approached in the right way.  
> For example, Nike golf shirts now have SPF 30 protection.  What types of
> questions did they ask consumers to come up with that idea?
>
> The product management team at Mozilla is currently working on
> mechanisms that will allow us to better capture user data to uncover
> these hidden stores and opportunities.  Expect to see some things roll
> out in the next month or so.

This sounds very good! I can only echo pascal here in asking you to make
sure you're very open about it.

> Cheers,
> Sherman

:)

--
dolphinling
<http://dolphinling.net/>
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3 (consumers vs enterprise)

Michael Lefevre
In reply to this post by pascal chevrel
["Followup-To:" header set to mozilla.dev.apps.firefox.]
... because there seem to be more posts in this group and crossposting
sucks.

On 2006-06-22, pascal <[hidden email]> wrote:
> Sherman Dickman a écrit :
>> The product management team at Mozilla is currently working on
>> mechanisms that will allow us to better capture user data
[...]
> - we need corporate deployment and *maintenance* tools to get into the
> enterprise. The CCK is a first small step, but the one major blocker for
> network administrators is maintenance. I have to say that I am quite
> concerned to see that all you said seems to be only based on the
> end-user and nothing is said about strategies for the corporate world.

I nearly followed up in bugzilla the other day on a discussion that
doesn't belong there (I decided against it and someone else did it
anyway... https://bugzilla.mozilla.org/show_bug.cgi?id=231062 )

Mike Connor wrote in
https://bugzilla.mozilla.org/show_bug.cgi?id=231062#c126 :
: We have consistently stated that our goal is to support consumers,
: enterprise deployment is an entirely different scenario that we're not
: set up to support at present.  Entering the enterprise market requires a
: lot more than just an MSI and some group policy support, its a matter of
: adopting an entirely different focus than targeting the consumer market.
: Right now, it doesn't make a lot of sense for us to do that.

I hunted around a little, and it's true that Mozilla Corporation stuff
recently has focused on consumers. Back in 2003/2004 though, there was
lots of talk of going for enterprises (as someone in the bug has quoted
Chris Hofmann saying so). More recent comments about enterprise have only
come from people of Mozilla Europe and Mozilla Foundation.

I should note that Mike Beltzner followed up in the MSI bug to say that
making it work was somewhere in middle of the list of things to do, and it
was just that "the value isn't worth blocking a release, nor has it gotten
to be a high point on anyone's radar. Perhaps the IBMers working on CCK
would like to take it or something like it. Perhaps you would like to
start building a list of the features that IT departments would truly
require, prioritizing the buglist, and driving the development of the
feature."

I guess at some point in 2005 it was decided that Mozilla's own
developers/marketing folks should focus on home users rather than try to
aim for enterprises as well. Was that a conscious decision or did it just
happen that way? Maybe it was at the point when MSI and stuff were dropped
from the list of blockers for Firefox 1.5?

--
Michael
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Jeff Walden-2
In reply to this post by pascal chevrel
pascal wrote:
>> * Which of our technology assets are clearly superior to the
>> competition?  How do we keep them from being easily copied or
>> duplicated?  Are there barriers to entry or significant hurdles for
>> competitors to overcome?
>
> I think I don't agree on this strategy.

Barriers to entry need not be artificial in nature.  For example, if another browser were to implement extensions it would have to implement something comparable to XUL with scriptability similar to that provided by JavaScript, because the code compilation step is a significant hurdle for most of the population of prospective extension developers.  Implementing a new version of XUL, embedding some sort of scripting with it, and making the underlying parts of your application accessible from script is a lot of work.  Consequently, extensions as a feature cannot be easily copied or duplicated -- not for anything approaching anti-competitive reasons, but simply because it's a whole lot of work to do it correctly.  Is it unfair to capitalize on the advantages extensions provide?

Jeff

--
Rediscover the Web!
http://snurl.com/get_firefox

Reclaim Your Inbox!
http://snurl.com/get_thunderbird
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Ben Goodger
Jeff Walden wrote:
> Barriers to entry need not be artificial in nature.  For example, if
> another browser were to implement extensions it would have to implement
> something comparable to XUL with scriptability similar to that provided
> by JavaScript, because the code compilation step is a significant hurdle
> for most of the population of prospective extension developers.  

For existing extension developers, perhaps.

I don't think you'd need something exactly like XUL though, and you
probably wouldn't want it. Mozilla's "infinite unfrozen API" (the user
interface) is the Achilles heel of the extension system. Version
checking and disabling mitigates the user's browser getting broken, but
it does not a good user experience make.

-Ben
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Mike Shaver
On 6/22/06, Ben Goodger <[hidden email]> wrote:

> Jeff Walden wrote:
> > Barriers to entry need not be artificial in nature.  For example, if
> > another browser were to implement extensions it would have to implement
> > something comparable to XUL with scriptability similar to that provided
> > by JavaScript, because the code compilation step is a significant hurdle
> > for most of the population of prospective extension developers.
>
> For existing extension developers, perhaps.
>
> I don't think you'd need something exactly like XUL though, and you
> probably wouldn't want it. Mozilla's "infinite unfrozen API" (the user
> interface) is the Achilles heel of the extension system. Version
> checking and disabling mitigates the user's browser getting broken, but
> it does not a good user experience make.

(Veering into dev-platform territory; I should make a post there about
this anyway.)

I think that you would very much want something like *XUL*, in that
the DOM-based declarative markup model for user interface operation
has been a pretty big success.  For all its many warts -- oh, XUL',
where art thou? -- I think it was a huge part of our ability to
deliver the applications we've delivered, and for extension developers
to be as productive and prolific as they've been.  (And it gave us a
straightforward way to drive the interface from script; writing
interface logic in native code, especially our 1998-era dialect of
C++, would probably have killed us in our tracks.)

You would probably want the overlay capability too.  What you don't
want, necessarily, is for people to be basically forced into building
dependencies on interface-implementation details.  bz and I talked a
bit last year about the idea of XBL widgets exposing "overlay targets"
or some other synthetic DOM for people to target, such that the real
implementation of the menu/widget/toolbar/etc. could be changed while
still providing support for some set of "frozen" integration points.

We would also benefit from some additional declarative support for
other common tasks, from "run this code on first run after extension
install" to "tell me when I've been added to the toolbar" to "notify
the user according to their preferences for this severity of datum",
because it would reduce the propagation of "just so" code from one
extension to another, and would give us again more flexibility in
adapting our foundation without disturbing the parts that extensions
rely on.  I'd be very interested in seeing more research and work done
along the lines of the search plugin format -- having the suggest URL
there is a vast improvement over people having to do extensions and
customize the widget somehow, *shudder* -- or greasemonkey, to capture
other common sorts of extensions.  I think this would also have
positive effects on security and user confidence, and would probably
let us give the user better guidance about the likely and possible
effects of their choices.

(I never did get to give my "the app is the api: surviving and
thriving with extremely extensible software" talk at ETech the other
year.  I should try to find some time to put that sort of thing
together as a presentation or paper or blog post or something...)

Mike
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Gervase Markham
In reply to this post by pascal chevrel
pascal wrote:
> - we don't have any market share in arabic countries because there is no
> official arabic mozilla portal nor any arabic online resources about
> mozilla.

What makes you think those are the two main reasons?

> - we need corporate deployment and *maintenance* tools to get into the
> enterprise.

Why is getting into the enterprise a goal?

> - native ODF compatibility would be a corporate killer feature for FF3

There's already an extension which allows viewing of ODF documents
directly in Firefox. And it's pretty simple. I certainly think we should
consider doing this for Firefox 3, because increased use of ODF furthers
the Mozilla Foundation's meta-goal of promoting choice and innovation on
the Internet.

Gerv
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Gervase Markham
Gervase Markham wrote:
 > There's already an extension which allows viewing of ODF documents
> directly in Firefox. And it's pretty simple. I certainly think we should
> consider doing this for Firefox 3, because increased use of ODF furthers
> the Mozilla Foundation's meta-goal of promoting choice and innovation on
> the Internet.

And, I should add, because the lack of easy availability of an ODF
viewer hinders adoption, because you can't email people ODF documents
and expect that they'll be able to read them.

Gerv
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

wyrfel
In reply to this post by Jeff Walden-2
Hi,

> I didn't test myself but from the descriptions I have read the Opera 9
> "Widget" system is already something similar.

No it isn't. It is rather something similar to GDesklets or dotWidget.
At least to my understanding.

André.

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Adam Hauner
In reply to this post by Gervase Markham
Gervase Markham wrote:

>> - we need corporate deployment and *maintenance* tools to get into the
>> enterprise.
>
> Why is getting into the enterprise a goal?

Tools for large-scale deployment and maintenance are keys for next
increase of market share for their browsers. Probably MoFo, MoCo and
whole community wanna see Gecko-based browsers on more computers.

Futhermore, the mission of the Mozilla project is to preserve choice on
the Internet. Big companies and organizations actually haven't the
choice of web browser because only one (?) browser has tools their need.

Network administrators still continue to ask and Mozilla project still
haven't satisfying answer.

Best regards
--
Adam Hauner
Projekt CZilla
http://www.czilla.cz/ - http://start.czilla.cz/
http://firefox.czilla.cz/ - http://thunderbird.czilla.cz/
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Michael Kaply-2
In reply to this post by pascal chevrel
pascal wrote:
> - we need corporate deployment and *maintenance* tools to get into the
> enterprise. The CCK is a first small step, but the one major blocker for
> network administrators is maintenance. I have to say that I am quite
> concerned to see that all you said seems to be only based on the
> end-user and nothing is said about strategies for the corporate world.

Can you please be more specific as to what you mean by "maintenance"

Mike Kaply
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Michael Kaply-2
In reply to this post by Adam Hauner
Adam Hauner wrote:

> Gervase Markham wrote:
>
>>> - we need corporate deployment and *maintenance* tools to get into the
>>> enterprise.
>>
>> Why is getting into the enterprise a goal?
>
> Tools for large-scale deployment and maintenance are keys for next
> increase of market share for their browsers. Probably MoFo, MoCo and
> whole community wanna see Gecko-based browsers on more computers.
>
> Futhermore, the mission of the Mozilla project is to preserve choice on
> the Internet. Big companies and organizations actually haven't the
> choice of web browser because only one (?) browser has tools their need.
>
> Network administrators still continue to ask and Mozilla project still
> haven't satisfying answer.
>
> Best regards

Are there bugs around this? Is MSI really all we're talking about here?

Is it all about Windows deployments?

Mike Kaply
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Ben Goodger
Michael Kaply wrote:
> Are there bugs around this? Is MSI really all we're talking about here?
>
> Is it all about Windows deployments?

For some deployments yes. Consider this. A company has a large number of
Linux and Windows machines. The Linux machines are used largely by the
engineers and the Windows machines by both engineers and non-engineers.

On the Linux side, there might be some support infrastructure in place
to manage Firefox, but it's not as important since the engineers can and
will configure things how they want.

On the Windows side, you need to support the less technical types, so
you will want/need a comprehensive automated installation/update mechanism.

People are crying out for MSI. The solutions at present are much more
complicated than an administrator that wasn't necessarily determined to
deploy Firefox are willing to put up with.

-Ben
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Mike Beltzner
On 23-Jun-06, at 3:57 PM, Ben Goodger wrote:

> Michael Kaply wrote:
>> Are there bugs around this? Is MSI really all we're talking about  
>> here?
>> Is it all about Windows deployments?

Bug 231062 has a lot of comments on this, although I'd probably start  
reading from the end. As I stated in comment 132, I don't think we  
have a good sense of what the requirements here are beyond "make an  
MSI", and I have a hard time believing that's all that an enterprise  
deployment would require. Maybe I'm wrong on that. The point is,  
nobody (to my knowledge) has really ever done a requirements analysis  
of things.

Mike Connor also mentioned (in this thread? I forget ...) that we  
want to be careful not to open a pandora's box of expectations here.  
If we provide an MSI, the next thing we'll be asked for is support  
for a given product version for longer than we might be capable of  
supporting it. Or maybe support for group policies. And if we don't  
provide those things because of the resource drain, does that  
invalidate all the work on MSI? I'm not saying it would, I'm saying  
we're not sure.

> People are crying out for MSI. The solutions at present are much  
> more complicated than an administrator that wasn't necessarily  
> determined to deploy Firefox are willing to put up with.

Would MSI be the only solution? If so, is it just a matter of  
building MSI support into our build process? Should we fork this thread?

cheers,
mike

_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Axel Hecht
In reply to this post by Ben Goodger
Mike Beltzner wrote:

> On 23-Jun-06, at 3:57 PM, Ben Goodger wrote:
>
>> Michael Kaply wrote:
>>> Are there bugs around this? Is MSI really all we're talking about here?
>>> Is it all about Windows deployments?
>
> Bug 231062 has a lot of comments on this, although I'd probably start
> reading from the end. As I stated in comment 132, I don't think we have
> a good sense of what the requirements here are beyond "make an MSI", and
> I have a hard time believing that's all that an enterprise deployment
> would require. Maybe I'm wrong on that. The point is, nobody (to my
> knowledge) has really ever done a requirements analysis of things.

I wouldn't claim "analysis", but here is what I have:

I talked to a T-Systems sales guy at CeBIT who is doing managed
desktops. I wanted to know what they say (or how loud they laugh) if a
client actually wanted Firefox or Thunderbird on their boxes. The bottom
line reply was "huh? You'd need more market share". "We're at 20-30% in
Germany." "You'd need a success story at least. How about the public
sector, those guys are into OpenSource, our guys are over there...". I
queued there for five minutes, but I had to leave after that.
Luckily, I was standing at the OpenOffice booth right at the time when
someone from the public sector came in and handed in his success story.
It only took them one year from success to story, due to policies and
lack of priorities.
Form feed, next fair, Linuxtag. There is a company doing corporate
deployment of Mozilla's in the public sector. Or did, the client was
voted out of office, the new government is doing different decisions.

Going for the corporate market is a multi-year game, and an MSI is cute
and all, but it's not the key. Not having one may be a showstopper, though.

> Mike Connor also mentioned (in this thread? I forget ...) that we want
> to be careful not to open a pandora's box of expectations here. If we
> provide an MSI, the next thing we'll be asked for is support for a given
> product version for longer than we might be capable of supporting it. Or
> maybe support for group policies. And if we don't provide those things
> because of the resource drain, does that invalidate all the work on MSI?
> I'm not saying it would, I'm saying we're not sure.

Ask 3 admins, get five opinions. And some of them are even bad ones. I
did, actually, CeBIT one year back, and at least one admin had group
policies and all kind of customization requests.

But that'd be actually corpfox, not firefox.

I don't think that listening to admin requests is going to be answer,
just as listening to user requests is not the answer for an end user
product.

Regarding the end-of-life problem, yes, that's going to be severe.
Another problem is that many of our services, in particular update and
addons, would be something that IT departments could possibly want to
include into their deployment and certification strategy, which would
require us to actually think about the license of that server code again.
With my business-me shining through, a corporate focus may require a
parallel business model, and may be a business opportunity for someone
not directly involved in the Mozilla Corporation itself. Whether the
core development community decides to not break those business models,
though, is a totally different question. There's this "let's remove the
pref, localizers shouldn't tweak it, users shouldn't tweak it" and then
mkaply comes and needs it in CCK. The architecture requirements are
still there, but that doesn't mean that mozilla.org or mozilla.com need
to commit to do the work.

>> People are crying out for MSI. The solutions at present are much more
>> complicated than an administrator that wasn't necessarily determined
>> to deploy Firefox are willing to put up with.
>
> Would MSI be the only solution? If so, is it just a matter of building
> MSI support into our build process? Should we fork this thread?
>

"Far from", and "maybe", resp.

Axel
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Robert O'Callahan-3
In reply to this post by Gervase Markham
Gervase Markham wrote:
> Why is getting into the enterprise a goal?

It should be a goal *eventually* because people surf the Web from their
offices a lot. Whether it makes sense to invest in this right now is not
so clear.

Having said that, companies like Novell that are pushing "enterprise
desktop Linux" are invariably implicitly pushing Firefox as well. Work
on Firefox manageability etc is being done, but not necessarily on Windows.

Rob
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

pascal chevrel
In reply to this post by Gervase Markham
Le 23/06/2006 12:54, Gervase Markham a ecrit :
> pascal wrote:
>> - we don't have any market share in arabic countries because there is no
>> official arabic mozilla portal nor any arabic online resources about
>> mozilla.
>
> What makes you think those are the two main reasons?

Because if you look at the languages/countries where we are strong,
these are the languages where we have end user support and online
presence, that's the chicken and egg problem. Even when a localized
version  doesn't exists many people use it in English because they know
that in case of doubt they can have help forums in their language and a
minimum set of information in their language, having official pages in
your languages also says to these people that they are not second-class
net-citizens, that the project is also for them even though their
community isn't as organized yet as the western one. Of course there are
other reasons to poor market share like gecko CSS rendering bugs in RTL
languages, but if I am arabic and see that there isn't even a download
page in my language, I will probably not even consider downloading it.


>
>> - we need corporate deployment and *maintenance* tools to get into the
>> enterprise.
>
> Why is getting into the enterprise a goal?

* Because convincing one system administrator of installing Firefox on a
5000 seats company means convincing a few people, while convincing 5000
individual users with lots of different motivations is harder
* Because lots of people all over the world only have web access at
their office and not at home, specially in poorer countries where
internet access is very expensive for an individual
* Because lots of web sites and software targetting the enterprise world
are IE-only just because we don't have a significant corporate market
share, that's bad for open standards defense
* Because there is a big demand for that, especially among smaller
organisations who don't have the ressources to create their own homemade
deployment and patching solutions and have similar needs to end-users.
Those people would like to switch to Firefox but can't
* Because people use at home what they use at work (MSoffice...)
* Because people also work remotely and are forced to use IE at home
even if they don't want to because of the extranet being IE only
* Because it's good for Mozilla as a platform
* Because of IE7 that will also ship with Vista upgrades in enterprise
deployments, i'd rather see people keeping their XP and installing
Firefox than Microsoft using the new IE7 capabilities as a big selling
point for Vista
* because of Thunderbird ligthing and it's shared calendaring
capabilities which is clearly targetting a networked environment
* Because when i say Corporate i also think administration, I know that
maintenance difficulty (particulary distant patching) is seen as a big
"con" for FF/TB in several big French administrations that have already
moved to Oo and would like to go further.

If Firefox/Thunderbird at the entreprise/university/administration is no
longer an objective and we only focus on the home user, actually I'd
like to know, because I receive a question about it almost everyday. In
this case I would just put a faq item online saying that deployment
tools aren't even considered.

Pascal
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

Axel Hecht-2
pascal chevrel wrote:

> Le 23/06/2006 12:54, Gervase Markham a ecrit :
>> pascal wrote:
>>> - we don't have any market share in arabic countries because there is no
>>> official arabic mozilla portal nor any arabic online resources about
>>> mozilla.
>>
>> What makes you think those are the two main reasons?
>
> Because if you look at the languages/countries where we are strong,
> these are the languages where we have end user support and online
> presence, that's the chicken and egg problem. Even when a localized
> version  doesn't exists many people use it in English because they know
> that in case of doubt they can have help forums in their language and a
> minimum set of information in their language, having official pages in
> your languages also says to these people that they are not second-class
> net-citizens, that the project is also for them even though their
> community isn't as organized yet as the western one. Of course there are
> other reasons to poor market share like gecko CSS rendering bugs in RTL
> languages, but if I am arabic and see that there isn't even a download
> page in my language, I will probably not even consider downloading it.


I don't buy that. I doubt that anybody can set up a community site
without a community. And we won't be able to find the right way to
approach a arab community from a cultural point of view anyway.

>
>>
>>> - we need corporate deployment and *maintenance* tools to get into the
>>> enterprise.
>>
>> Why is getting into the enterprise a goal?
>
> * Because convincing one system administrator of installing Firefox on a
> 5000 seats company means convincing a few people, while convincing 5000
> individual users with lots of different motivations is harder

I doubt you need to convince one sysop, but all the management to not
fire him for that.

> * Because lots of people all over the world only have web access at
> their office and not at home, specially in poorer countries where
> internet access is very expensive for an individual

I don't think that that's a market we really need to target. Given all
the other users we don't have converted yet.

> * Because lots of web sites and software targetting the enterprise world
> are IE-only just because we don't have a significant corporate market
> share, that's bad for open standards defense
> * Because there is a big demand for that, especially among smaller
> organisations who don't have the ressources to create their own homemade
> deployment and patching solutions and have similar needs to end-users.
> Those people would like to switch to Firefox but can't

The nice thing is, we had a breakout session on the topic at the onsite
yesterday, and it seems that we're in much better shape than almost
anybody in the room thought.
See http://wiki.mozilla.org/AllHands/2Q2006/Corporate_Deployment for
more, but the main piece is Mission Control Desktop, and we have
volunteers to add docs to devmo for that.

> * Because people use at home what they use at work (MSoffice...)
> * Because people also work remotely and are forced to use IE at home
> even if they don't want to because of the extranet being IE only
> * Because it's good for Mozilla as a platform
> * Because of IE7 that will also ship with Vista upgrades in enterprise
> deployments, i'd rather see people keeping their XP and installing
> Firefox than Microsoft using the new IE7 capabilities as a big selling
> point for Vista
> * because of Thunderbird ligthing and it's shared calendaring
> capabilities which is clearly targetting a networked environment
> * Because when i say Corporate i also think administration, I know that
> maintenance difficulty (particulary distant patching) is seen as a big
> "con" for FF/TB in several big French administrations that have already
> moved to Oo and would like to go further.
>
> If Firefox/Thunderbird at the entreprise/university/administration is no
> longer an objective and we only focus on the home user, actually I'd
> like to know, because I receive a question about it almost everyday. In
> this case I would just put a faq item online saying that deployment
> tools aren't even considered.

The general idea as I understood it in the meeting is that we don't have
the resources to make a commitment to corporate needs. We do want to set
up a central point for interested parties to coordinate and communicate
and share their experience. If that leads to folks contribute patches,
those will be luckily taken, but we're not going to make market analysis
or dedicate management folks to any tasks in the area.
 From an engineering point of view, I think that most of use
underestimated by far how well things can work, and we'll try to convey
that message a bit better overall.

Axel
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
Reply | Threaded
Open this post in threaded view
|

Re: PM Activities for Firefox 3

pascal chevrel
Axel Hecht a écrit :

> pascal chevrel wrote:
>> Le 23/06/2006 12:54, Gervase Markham a ecrit :
>>> pascal wrote:
>>>> - we don't have any market share in arabic countries because there
>>>> is no
>>>> official arabic mozilla portal nor any arabic online resources about
>>>> mozilla.
>>>
>>> What makes you think those are the two main reasons?
>>
>> Because if you look at the languages/countries where we are strong,
>> these are the languages where we have end user support and online
>> presence, that's the chicken and egg problem. Even when a localized
>> version  doesn't exists many people use it in English because they
>> know that in case of doubt they can have help forums in their language
>> and a minimum set of information in their language, having official
>> pages in your languages also says to these people that they are not
>> second-class net-citizens, that the project is also for them even
>> though their community isn't as organized yet as the western one. Of
>> course there are other reasons to poor market share like gecko CSS
>> rendering bugs in RTL languages, but if I am arabic and see that there
>> isn't even a download page in my language, I will probably not even
>> consider downloading it.
>
>
> I don't buy that. I doubt that anybody can set up a community site
> without a community. And we won't be able to find the right way to
> approach a arab community from a cultural point of view anyway.

You think that there was a community when I created the first Spanish
web forums years ago? There wasn't any at this time and I started alone,
it worked. As for the cultural point of view, we obviously don't live in
the same culture since I even started learning arabic at university
which is pretty common in France where you can start arabic lessons in
many secundary schools as a second language and where the aïd el kebir
is almost as celebrated as christmas...


>
>>
>>>
>>>> - we need corporate deployment and *maintenance* tools to get into the
>>>> enterprise.
>>>
>>> Why is getting into the enterprise a goal?
>>
>> * Because convincing one system administrator of installing Firefox on
>> a 5000 seats company means convincing a few people, while convincing
>> 5000 individual users with lots of different motivations is harder
>
> I doubt you need to convince one sysop, but all the management to not
> fire him for that.

Please, try to open your mind just as second and stop taking points at
first degree. The point is that big chunks of the browser market do not
depend *at all* on the end-user, they are not those who control the
software on their machines, therefore either we decide that large chunks
of the browser market are IE only, or we don't.

>
>> * Because lots of people all over the world only have web access at
>> their office and not at home, specially in poorer countries where
>> internet access is very expensive for an individual
>
> I don't think that that's a market we really need to target. Given all
> the other users we don't have converted yet.

Strange reasonning, as if some users would explude others...

>
>> * Because lots of web sites and software targetting the enterprise
>> world are IE-only just because we don't have a significant corporate
>> market share, that's bad for open standards defense
>> * Because there is a big demand for that, especially among smaller
>> organisations who don't have the ressources to create their own
>> homemade deployment and patching solutions and have similar needs to
>> end-users. Those people would like to switch to Firefox but can't
>
> The nice thing is, we had a breakout session on the topic at the onsite
> yesterday, and it seems that we're in much better shape than almost
> anybody in the room thought.
> See http://wiki.mozilla.org/AllHands/2Q2006/Corporate_Deployment for
> more, but the main piece is Mission Control Desktop, and we have
> volunteers to add docs to devmo for that.


Good news.

>
> The general idea as I understood it in the meeting is that we don't have
> the resources to make a commitment to corporate needs. We do want to set
> up a central point for interested parties to coordinate and communicate
> and share their experience. If that leads to folks contribute patches,
> those will be luckily taken, but we're not going to make market analysis
> or dedicate management folks to any tasks in the area.
>  From an engineering point of view, I think that most of use
> underestimated by far how well things can work, and we'll try to convey
> that message a bit better overall.

Thats all I was asking, thanks.

Pascal
_______________________________________________
dev-planning mailing list
[hidden email]
https://lists.mozilla.org/listinfo/dev-planning
12