curriculum-foundation icon indicating copy to clipboard operation
curriculum-foundation copied to clipboard

Provide proposals what to cut if training last only 3 days instead of 4

Open sippsack opened this issue 2 years ago • 22 comments

There are different approaches how to handle all the contents of the curriculum if the training last only 3 days:

  • R1 topics are required, R2/R3 is optional and the attendees can choose
  • send a reading guide/recommendations to the attendees in advance and treat some of the topics shorter
  • ...

Idea: provide different variations as proposals to help trainers by deciding, what to cut

This must not be part of the curriculum, but maybe a blog post or a info at the iSAQB website

sippsack avatar May 25 '22 10:05 sippsack

@sippsack could you come up with a list of proposals for

  • topics that could be removed from the curriculum
  • topics that could be downgraded from R1/R2 to R3, therefore becoming purely optional

gernotstarke avatar Oct 24 '22 15:10 gernotstarke

after discussion between GS and AL we decided to remove this from the V2023 version.

@sippsack - please provide a wording-proposal for a blogpost, as we currently don't have any other medium where we could publish this.

gernotstarke avatar Oct 26 '22 19:10 gernotstarke

My personal candidates for removal or setting to R3 are as follows.

  1. LG 1-3: Understand software architecture as part of the software lifecycle (R2)
  2. LG 1-6: Can explain the correlation between development approaches and software architecture (R1) - Covered in advanced classes
  3. LG 1-7: Differentiate between short- and long-term goals (R1)
  4. LG 2-1: top-down and bottom-up approaches to design (R1)
  5. LG 2-3: Identify and consider factors influencing software architecture (R1-R3) - make more general
  6. LG 3-1: Explain and consider the quality of technical documentation (R1)
  7. LG 3-9: Use documentation as written communication (R2)

I'm not saying that these are unimportant, but some are covered in advanced classes and some are simply less important than the other learning goals.

rhoadesre avatar Oct 30 '22 14:10 rhoadesre

  • LG 1-3 has been discussed in #252 (and we decided to leave it as-is)

  • LG 1-7 has already been adressed in #251

  • reducing LG 1-6 to R2 gets a +1 from me.

Personally I consider 2-1, most of 2-3, 3-1 and 3-9 actual core competencies of software architects, and would really like to have those remain R1.

I really doubt that we will come with a consensus on importance of topics, I consider the different customers and projects we're working in the source of that "evil": Most of the stuff in our curriculum can be important for some project or team...

Maybe we should keep the ultra-tight schedule - it has worked out for several years now...

gernotstarke avatar Oct 30 '22 15:10 gernotstarke

Personally I consider 2-1, most of 2-3, 3-1 and 3-9 actual core competencies of software architects, and would really like to have those remain R1.

To be clear, I also consider LG 2-1 and 2-3 core competencies and would leave them as R1. I would just reduce the number of R1 topics. For example, as part of LG 2-1, I was only referring changing "top-down/bottom-up approach" to R3. I would leave the rest of 2-1 I as is. I would also generally leave LG 2-3 as R1, but make it more general and not have so many R1 sub-topics.

LG 3-1 is important, just not as important as LG 3-2. I also emphasize (a lot!) that "form, content, and level of detail [should be] tailored to the stakeholders" in my classes. However, I still consider it a candidate for something that can be demoted to R3.

LG 3-9: Use documentation as written communication is simply too general for me and is already covered by LG 3-2. How would you create a test question for this blanket statement?

rhoadesre avatar Oct 31 '22 07:10 rhoadesre

What about making LG 4-3 and LG 4-4 both R3 only, as "analysis and evaluation" is covered in AWERT and IMPROVE...

most junior architects will not evaluate systems on their own, so imho we could safely defer these topics to the advanced level.

gernotstarke avatar Nov 09 '22 19:11 gernotstarke

We should cover in the foundation level the topics that will be asked in any advanced exam, regardless of the modules attended! Example: Quality tree and quality scenarios.

mahboubagharbi avatar Nov 09 '22 19:11 mahboubagharbi

Torn between to sides... either

  1. shorten the curriculum, and reduce stress on foundation trainings OR
  2. prepare people for advanced level.

My current opinion is 1 - as only less than 10% of people go for advanced, AND I can expect advanced architects to do some research on required topics.

Plus - quality tree and scenarios are covered in LG 4-2.

gernotstarke avatar Nov 09 '22 19:11 gernotstarke

This was just an indication of when to choose between topics. Some examiners point out foundation level topics during the interview. Regardless: I think feedback from the exams sub-working group on the curriculum would be valuable.

mahboubagharbi avatar Nov 09 '22 20:11 mahboubagharbi

I would like to keep LG 4-3 as (partly) R2, as it currently is. Since we stress the importance of architecturally significant requirements and constraints, participants should have at lease some idea who to evaluate the architecture in that regard.

ulibecker avatar Nov 09 '22 20:11 ulibecker

In addition, you can get advanced-level certified without attending AWERT and IMPROVE. An advanced-level certified architect should have learned at least the foundations of architecture evaluation.

ulibecker avatar Nov 10 '22 07:11 ulibecker

As I mentioned above, I'm clearly torn between two sides... In advanced exams, people have to know more than what's covered in FL, e.g. cloud scaling in some exams.

The foundation should NOT be considered sufficient to pass an advanced exam - but what the title says - a foundation where additional knowledge can be built upon.

We should have a working-group meeting for that discussion!!

gernotstarke avatar Nov 10 '22 07:11 gernotstarke

I'm also in favour of a working group meeting to discuss this.

However, I also think that analysis/assessment of software architecture should remain in the curriculum. People might never get in touch with that topic in their day-to-day job, so at least pointing them towards that area should be part of the curriculum. Doesn't have to be ATAM, can be Mini quality attribute workshop, or just the basic concept of evaluating an architecture based on a few aspects.

programming-wolf avatar Nov 10 '22 07:11 programming-wolf

I suggest less design patterns but more focus on the architecture themes. I believe that the themes:

LG 1-3: Understand software architecture as part of the software lifecycle (R2)
LG 2-1: top-down and bottom-up approaches to design (R1)
LG 2-3: Identify and consider factors influencing software architecture (R1-R3) 
LG 3-9: Use documentation as written communication (R2)

Absolutely should remain in the curriculum.

Candidates for removal or setting to R3 are as follows: LG 1-6: Can explain the correlation between development approaches and software architecture (R1) LG 1-7: Differentiate between short- and long-term goals (R1) LG 3-1: Explain and consider the quality of technical documentation (R1)

I am in favor of a streamlined curriculum that includes more time for hands-on group work. The participants would like to experience more practice than theory in the training. Most have already read many books.

mahboubagharbi avatar Nov 10 '22 09:11 mahboubagharbi

It seems we don't come to definitive conclusions here in the written discussion. Let's make this a topic on a face2face working group meeting - but we won't manage that in time for the upcoming 2023 release (we need to freeze changes by the end of 2022 latest).

Therefore, I see the following options:

  1. decide in 2023 how and where to streamline, and have another release 2024 (breaking our 2-year release cycle)
  2. keep things for another two years, and postpone the streamlining to 2025.

I personally suggest to aim for 1), as it would help everybody to have less stuff to present.

gernotstarke avatar Nov 18 '22 09:11 gernotstarke

1. decide in 2023 how and where to streamline, and have another release 2024 (breaking our 2-year release cycle)

I'm all for that.

mikesperber avatar Nov 25 '22 13:11 mikesperber

Sorry, I'm late to this lively discussion. I didn't find any notifications in my inbox (maybe they have been in spam and are gone now). I have to check my github settings.

Thanks for all your suggestions. I'll try to consolidate them and come up with a wording proposal as Gernot asked for in October.

We can discuss it further in a face2face meeting. I am very curious to see if we can find an agreement.

sippsack avatar Mar 03 '23 21:03 sippsack

There won't be a face2face meeting in near future, I think. Therefore I will plan a few short remote meetings with two or three of you at a time to discuss all the suggestions. One remote meeting for all won't fit our timetables and may be difficult to moderate.

sippsack avatar Mar 05 '23 12:03 sippsack

Hello Falk,

The discussion was a while ago ;-). I now suggest we start with the new curriculum and gain experience with it. At the end of summer / beginning of autumn we should take up the topic again.

Regards, Mahbouba

mahboubagharbi avatar Mar 05 '23 14:03 mahboubagharbi

+1

gernotstarke avatar Mar 05 '23 14:03 gernotstarke

At least I get mail notifications for this issue now. I'll set a reminder in my calendar.

sippsack avatar Mar 06 '23 08:03 sippsack

+1

rhoadesre avatar Mar 06 '23 10:03 rhoadesre

I close this issue - as we shall not mix curriculum with training-organization:

In the curriculum, we concentrate ONLY on learning goals and terminology. We leave it to trainers and/or training organizations to map this curriculum on their training schedule - wether it's 3 days, 4 or even longer.

gernotstarke avatar Jul 24 '24 18:07 gernotstarke