So, squeezing the provider ad infinitum:
Can we get a low-cost product?
No, what we get is one less supplier.
Low cost is a digital workplace strategy that requires
absolutely adjust the business areas
so that everything fits
Our software manufacturing processes are continually being refined, refined, and streamlined. We rely on highly efficient ecosystems that ensure their economy, extensively detailed in our series on eco-logical manufacturing. Optimization efforts are not minor in Testing and Certification processes, being ours testing orientation towards factory models a clear proof of it. However, the strategy is missing:
Quality a priori before quality a posteriori
- Build on a predefined and guaranteed service architecture.
- Consider software as an accounting asset. Controlled, managed and audited.
- Set base patterns (as CRUD patterns).
- Standardize common services (such as traceability, validation or utilities).
- Segment in information contexts: application, session and operation.
- Separate operations into layers (presentation, business -for example-).
- Automate the generation of metrics and quality control on artifacts.
- Synchronize Software Quality Assurance (SQA) with the entire life cycle.
- Centralize artifacts (services) and reuse them.
- Train and share knowledge about centralized artifacts (services).
- Differentiate between the work of architects and that of programmers.
If the way we build, evolve, or test our software causes each cycle increases the workload (code, functional, regressive ... whatever we want), We will never achieve a low-cost software quality strategy. No matter how tight the cost wants to be, if we multiply it by infinity, the result will always be great.
To sleep peacefully, we will have to handle metrics that tell us that approximately 80% of the code is based on patterns or services that have already been tested and are reliable. AND IT IS POSSIBLE.
If, even, it is possible to develop a business of beautiful low-cost houses (The BoKlok concept), How can we not have a low-cost software quality strategy?
Yet another more extreme example of how to change the way you do things:
Finally, a little play on words.
There are low-cost elements for which our culture already has explicit names, which leave no room for imagination. The «low-cost jewels» are costume jewelery or trinkets, the «low-cost wine or drink» is fierce wine or carafe and the «low-cost accommodation» is a pension.
A great reflection on a magnificent post. Congratulations.
I think that this "low cost" that you mention, the result in part of this crisis, is nothing more than adjusting what you pay to what you really want, without additional easements or services that you have not asked for, and for which you do not want to pay.
I think that several of the points you make in your strategy go along this line: Try only what is necessary, but with the necessary means.
The price adjustment effect, typical of a "pay for what you consume" model, I consider innate to factory models.
However, even in factory models, a huge task will have a high cost / price. To ensure that Software Quality is kept at low costs, the volume of the task has to be contained and that (I think) is the key to this low-cost strategy.
In other words, objective: that the new deliveries generate a "small" volume of tests to be carried out.
(easy to say ...)
A brilliant post, as always 😉
I totally agree that the way of doing things must be changed. Pe, is today's marketing really low cost? I do not believe it. When marketing ran out of budgets, it had to reinvent itself, and the result has been a new way of marketing, cheaper, but also more imaginative and efficient. Just look at the layout of this post! ;-D
Thank you Lucia,
(sign up for both the layout 😉
As put in the article on the debate table of INNOVATION DMQuality2013:
"How do we do it? Putting it into practice. »
Shall we start with OGMA?