Re: [flexcoders] Re: **JOB **Flex developer needed ASAP (Contract to hire!!).

[Anatole Tartakovsky]

Jason, I would like to stress one point in Dave's comment that is crucial to success of RIA project. Those who been through client-server wave know that the process Dave describes is the best way to get useful application rather then just "working" one. The process starts with quick and recreation (few hours or days at most as starting point- static data) of the current process just to get consultants into domain. Their task is to introduce RIA to the current process and open the users to idea that anything is possible. The process of users actually being interviewed with working interactive prototypes being developed within hours of discussion renders current bureaucratic approach to design obsolete. Within week(s) the high-level specifications are developed as an interactive prototype. I would also note that as initial design is done it becomes weekly / biweekly reassessment process. 2 goals needed to be achieved : 1. Provide users with early access to modules that allow them to do pilot of functionality/QA/usability testing 2. Maintain realistic schedule - after the first week the users will assume that the project will be done by the end of the next week   The only thing I saw in the field that differs from Dave's assessment is the size of the team. I found it extremely helpful to maintain not 1-3 person but 3-5 people teams - usually by complementing our consultants/software engineers with client's developers/business/QA people. While it seemed in the beginning like waste of valuable time for unnecessary communications, it assures that the end-product is accepted and maintained by the client before the final crunch time - and also minimizes communications with end-users beyond technical aspects. It does add 50% overhead to the times Dave mentioned. Other (obvious) reason is that it takes time for client to realize that 40 people teams/outsourcing model have to change and the people need to be retrained for different process.   Thank you, Anatole   ----- Original Message ----- From: Dave Wolf To: flexcoders@yahoogroups.com Sent: Thursday, April 13, 2006 5:46 PM Subject: [flexcoders] Re: **JOB **Flex developer needed ASAP (Contract to hire!!). Hey Jason, I am going to try my best to inline my answers here. --- In flexcoders@yahoogroups.com, "Jason Hawryluk" <[EMAIL PROTECTED]> wrote: > > Dave, I fully realize the potential of Flex to let you crank out an > application quickly. You have a "Team" of people this is a requirement for > "one" contractor. The thing that got me was the �rollout� part of the > demand. > You're right.  We do have "teams" here at Cynergy.  I think its worth noting though that in the applications I am discussing, these teams have ranged from 1 to 3 people.  I've seen posting in the past where people talk about their projects with 30-40 devs.  I cannot honestly fathom what problem set they're trying to solve. > > > I agree that when you fully understand the domain of the intended �user� of > the product then yes that is quite possible. Having no idea of the user > base, and their goals, infrastructure, budgets, requirements, and resources, > how is this assumption possible? > > It's of course not.  The key is to develop a process by which you can extract those requirements as a part of the development process itself, rather then some incredibly extended requirements gathering phase.  First off, the longer that phase takes the less relevant the data you collect actually becomes.  Secondly dependant on how you collect it, (traditional JAD approach) the value of what you learn can be pretty suspect. > > Cranking out a product that fit�s someone�s specifications, and creating a > tool that is goal/task oriented, and helps the �user� accomplish what they > need/want in a coherent fashion, are 2 different things. > > Could't agree more.  I think if you interviewed the folks we've done this development for, they'd tell you what they got was the latter. > > A certain percentage of the time required before any code is written is > spent interviewing management, users, understanding the goals and tasks that > a product needs to fulfill. > > Our approach here at Cynergy is to not view these as distinct phases. Rather to incorporate the domain experts into the development process from moment one, in the most transparant way possible.  We accomplish this by a development approach we call LookFirst, but really is a "front to back" development approach whereby the requirements gathering and "interviews" are done in real time with the development of the user experience.  The actual creaative act of development is used as a requirements gathering process. BTW at this point I should mention that we are currently writing a series for Adobe's DevNet where we will go through our approach and how to use it yourself. > > A software program should not be something that enforces a work method, and > because the person usually in charge of writing specs is IT (more often then > not, that person has not done the necessary foot work). What I mean is, you > end up with a product that just fills a business need from a management > perspective, and does not take into account segmented departmental user > goals. > > I contend a long interview and requirements gathering phase does nothing to alleviate this.   There is nothing intrinsic in the JAD process that fixes this shortcoming. > > So, sure cranking out data entry forms, or lists of data that fulfill a > business �need� is rapid. However, is that a solution that will help the > user do their job? Will that solution allow different user types > (commercial, management, technical, administrative) to fulfill their role in > the organization as efficiently as possible? Will the program have a memory > for example so it can make assumptions based on past user interaction? > > Well, if I had assumed we were talking about simple CRUD screens I would have said hours or days, not weeks for a prototype.... As I mentioned before, from moment one the domain experts are sitting right next to an interactive designer, a RIA developer and a data modeller.  They talk to and listen to the domain experts and in real time, right in front of them start putting together that user experience. If you want to see a domain experts eyes light up like a kid at a candy store, turn their wants and desires into something they can see and touch right in front of them.  They will become animated, interested, involved and will provide you a level of information and requirements you couldnt have extracted in months. > > I don�t believe in creating a tool that �only� fulfils the business > requirements from a management perspective. The users know how to do their > job. I would rather take the time to confirm that the product will meet > their goals and tasks, and give them a product that helps them do it better > based on �real� information. Management rarely understands this. Software > has become a necessary evil that enforces uneducated assumptions on the > user, instead of a pleasant, and comfortable tool that �helps� a user do > their job. > > As do we.  Thats the whole point of this approach. > > So while you may be able to do this in 6 months rollout, and all. I don�t > believe one person can do this �correctly� in 6 months. Now of course that > all depends on the size of the application, of which we have no idea. > > Prototypes in weeks? Whole business in 6 months? It takes weeks just to get > through the interview process, identify roles, identify goals for each role, > tasks for each goal, and confirming this research. > > Your internal processes, and goals are apparently quite different then mine. > Our goals are not.  The way we deliver those goals to our clients indeed are. I can only point to our successes as proof the approach works. Whether is obvious places you can look to like our MAX Award nomination last year, or the Fortune 100 clients we have done just this for in the last 4 months alone.  Just as the RIA has been a disruptive technology that will re-shape the way we look at web-based applications, so will the front-to-back approach be a disruptive influence that reshapes the way we develop them. We know we can deliver better software faster and at considerably less cost to our clients by working this way.  We know when we put the user experience first and foremost, usefulness follows naturally. -- Dave Wolf Cynergy Systems, Inc. Macromedia Flex Alliance Partner http://www.cynergysystems.com Email:  [EMAIL PROTECTED] Office: 866-CYNERGY > Jason Hawryluk > >  -----Message d'origine----- > De : flexcoders@yahoogroups.com [mailto:[EMAIL PROTECTED] la > part de Dave Wolf > Envoy� : mercredi 12 avril 2006 20:09 > � : flexcoders@yahoogroups.com > Objet : [flexcoders] Re: **JOB **Flex developer needed ASAP (Contract to > hire!!). > > > >   Off topic I guess, but six months for a prototype?  We've been rolling >   out pretty major applications in Flex from soup to nuts in six months. > >   To me all the sex appeal aside, some of the biggest value in Flex is >   what a productive canvas it is to work within.  Between Flex, great >   devs, good tools and frameworks, we're cranking out applications >   people are then running a whole business on in under six months. >   Prototypes.... we're talking weeks. > >   -- >   Dave Wolf >   Cynergy Systems, Inc. >   Macromedia Flex Alliance Partner >   http://www.cynergysystems.com > >   Email:  [EMAIL PROTECTED] >   Office: 866-CYNERGY > >   --- In flexcoders@yahoogroups.com, "Jason Hawryluk" <jh@> wrote: >   > >   > Sorry if this sounds harsh, or like a rant... >   > >   > >   > >   > You know I think about this kind of thing, and always wonder who in >   their >   > right mind consultant, or not would sign on to an already "presumed" >   6 month >   > project. Unless they have really done their homework (unlikely) how >   do they >   > know that 6 months is enough. >   > >   > >   > >   > So what they really should say is your going to develop this miracle >   > product, and oh it must be done in 6 months. Ready to deploy. Then they >   > plunk the old "needed ASAP" onto that. >   > >   > >   > >   > I smell disaster about to happen. >   > >   > >   > >   > If you read it it says prototype, ok that makes sense 6months for a >   > prototype no problem. Then the next phrase has "rollout"?. Then the big >   > hook. "Since client has already defined design specs". >   > >   > >   > >   > So to summarize they have a demand for a report generator >   "prototype" that >   > will be "rolled" out "live" in "6months" based on the "clients specs" >   > authored by an individual that most likely has no knowledge of Flex >   at all. >   > >   > >   > >   > I don't mean to be a pain but, could you please repeat the question? >   > >   > >   > >   > I just don't get it. Maybe it's just me. :) >   > >   > Jason >   >   -----Message d'origine----- >   >   De : flexcoders@yahoogroups.com >   [mailto:[EMAIL PROTECTED] la >   > part de wessam_jad >   >   Envoy� : mardi 11 avril 2006 20:01 >   >   � : flexcoders@yahoogroups.com >   >   Objet : [flexcoders] **JOB **Flex developer needed ASAP (Contract to >   > hire!!). >   > >   > >   > >   > >   >   Position Needed: Flex Developer >   >   Duration:6 months contract to hire. >   > >   >   Role Description � Consultant will be responsible for creating a new >   >   reporting interface prototype using Macromedia Flex. Prototype will >   >   most likely be developed using Flex 1.5. Subsequent versions could >   >   be developed in Flex 2.0. Consultant will be responsible for >   >   upgrade, replication, and rollout. Consultant will also participate >   >   in daily knowledge transfer sessions. Since client has already >   >   defined design specs, consultant is generally expected to complete >   >   assigned development tasks. >   > >   >   Desired Skills: >   >   -6 months Macromedia Flex 1.5 / 2.0 development experience >   >   -data warehouse, business intelligence experience >   >   -Excellent communication skills >   > >   >   send your resume ASAP to wjad(at)esginc(dot)com >   >   www.esginc.com >   > >   > >   > >   > >   > >   > >   > >   > >   > >   > >   > >   >   -- >   >   Flexcoders Mailing List >   >   FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt >   >   Search Archives: >   http://www.mail-archive.com/flexcoders%40yahoogroups.com >   > >   > >   > >   > >  -------------------------------------------------------------------------- > -- >   > -- >   >   YAHOO! GROUPS LINKS >   > >   >     a..  Visit your group "flexcoders" on the web. >   > >   >     b..  To unsubscribe from this group, send an email to: >   >      [EMAIL PROTECTED] >   > >   >     c..  Your use of Yahoo! Groups is subject to the Yahoo! Terms of >   > Service. >   > >   > >   > >  -------------------------------------------------------------------------- > -- >   > -- >   > > > > > > > > >   -- >   Flexcoders Mailing List >   FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt >   Search Archives: http://www.mail-archive.com/flexcoders%40yahoogroups.com > > > > ---------------------------------------------------------------------------- > -- >   YAHOO! GROUPS LINKS > >     a..  Visit your group "flexcoders" on the web. > >     b..  To unsubscribe from this group, send an email to: >      [EMAIL PROTECTED] > >     c..  Your use of Yahoo! Groups is subject to the Yahoo! Terms of > Service. > > > ---------------------------------------------------------------------------- > -- > -- Flexcoders Mailing List FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt Search Archives: http://www.mail-archive.com/flexcoders%40yahoogroups.com YAHOO! GROUPS LINKS  Visit your group "flexcoders" on the web.    To unsubscribe from this group, send an email to:  [EMAIL PROTECTED]    Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.