Software Estimation Quotes

The only way to reduce the variability in the estimate is to reduce the variability in the project.

When executives ask for an “estimate,” they’re often asking for a commitment or for a plan to meet a target.

Technical staff were fairly good at estimation but were poor at defending their estimates

The creator of Bitcoin, Satoshi, disappeared back in 2011, leaving behind open source software that the users of Bitcoin could update and improve. Five years later, it was estimated that only 15 percent of the basic Bitcoin computer code was the same as what Satoshi had written.

Instead, they face heightened anxiety and sleep deprivation, which causes dramatic mood swings and is responsible for an estimated 13 percent of highway deaths. Worse yet, since the software is designed to save companies money, it often limits workers’ hours to fewer than thirty per week, so that they are not eligible for company health insurance.

One study estimates that by 2019 the average worker was sending and receiving 126 business emails per day, which works out to about one message every four minutes.2 A software company called RescueTime recently measured this behavior directly using time-tracking software and calculated that its users were checking email or instant messenger tools like Slack once every six minutes on average.3 A

Third, enumerating the frequencies explicitly helps everyone recognize which decisions depend upon which user set properties. Even this sort of informal sensitivity analysis is valuable. When it develops that very important decisions are hinging on some particular guess, then it is worth the cost to establish better estimates for that value.

Where each entry is akin to a mini project charter for that work package; including such information as a description of the work, [Cost] estimates, [Scope] requirements & technical references, acceptance criteria & [Quality] standards, [People] & [Resource] needs, assumptions & constraints, [Time] duration & milestones.

By combining software with another, more readily monetized product — services, in this case — Amazon is able to efficiently extract profit from a growing, volume market. What’s more impressive, however, is that because Amazon is building primarily from either free software (in the economic sense) or software it developed internally, it is paying out minimal premiums to third parties for the services it offers. Which means that not only is AWS a volume business, it may be a high-margin business at the same time. Amazon does not break out its AWS revenues, so we’re forced to rely on estimates, but UBS analysts Brian Fitzgerald and Brian Pitz projected in 2010 that AWS’s margins would grow from 47% in 2006 to 53% in 2014. Last year, Andreas Gauger, the chief marketing officer for Amazon competitor ProfitBricks, estimated Amazon’s margins were better than 80%.

¿cómo sabrás que ya está terminada? Usemos como ejemplo la historia de Stoll: Como médico de fuerzas especiales, debo enseñar fisiología básica a mis alumnos, para que conozcan el cuerpo humano. Yo uso siempre un recurso nemotécnico para saber si una historia está lista. Fue creado por Bill Wake, especialista en diseño de software. Wake dice que para que una historia esté lista debe cumplir los criterios INVEST (“Invierte”): Independiente. Debe ser practicable y completable por sí sola. No debe depender inherentemente de ninguna otra. Negociable. Hasta que no esté terminada, debe ser posible reescribirla, llevar integrada la tolerancia a cambios. Valiosa. Tiene que aportar valor a un cliente, usuario o interesado. Estimable. Debe ser posible evaluarla.

Second, our estimating techniques fallaciously confuse effort with progress, hiding the assumption that men and months are interchangeable.

Technology partners What kind of customers are you trying to attract? If you give API access, then you attract software developers. Can you provide the support that they need, and is it worth your time? Most companies that built for us were very small, and we failed to generate significant revenue from API access. Almost all companies use a commercial website, and custom websites are rare. Here are the pros and cons of technology partners: Pros: You place 3rd party developers on the hook for support and maintenance. You free up developer time. You can expand your customer base. You lack developers to connect to a 3rd party system. When I built a QuickBooks integration for Kentico CMS, I asked them why they never built one themselves. Their response was that QuickBooks was not their business model. Connecting to QuickBooks is challenging and it requires a heavy lift for software developers. Cons: Building an integration could take several hours. Instead of building API access, can they integrate with you in another way? We pull orders from a variety of 3rd party shipping tools. Can the customer pull their sales into the shipping tool? Some developers fail to properly maintain and support their plugin. Your customers will call you and ask your company for help. If the 3rd party fails to respond, then you are in trouble. I advise gating your developer API to legitimate software companies only. Your company must provide developer support, which is also expensive and time-consuming. We had several instances where companies required multiple calls. It is difficult for some 3rd parties to follow developer guides and estimate costs. The 3rd party may have few clients and the cost to onboard the developer exceeds the sales.

For estimating construction costs of above-ground elements, readily accessible guides are available. Examples include the annual RSMeans Building Construction Cost Data series and Sage Estimating software.

velocity is a relative and team-dependent measure, not an absolute one. Teams usually have significantly different contexts which render their velocities incommensurable. Second, when velocity is used as a productivity measure, teams inevitably work to game their velocity. They inflate their estimates and focus on completing as many stories as possible at the expense of collaboration with other teams (which might decrease their velocity and increase the other team’s velocity, making them look bad). Not

The process is called estimation, not exactimation. —Phillip Armour

According to Section 2.4.2.3 of the PMBOK® Guide, iterative and incremental life cycles are those in which the project scope is generally determined early in the project life cycle, but time and cost estimates are routinely modified as the project team's understanding of the product increases.

Well-Estimated Project If the estimation inputs and process are well-defined, arbitrarily changing the output is not a rational action. Project stakeholders might not like the output, but the appropriate corrective action is to adjust the inputs (for example, reduce the project’s scope) and to recalculate the outputs, not just to change the output to a different answer.

Recommended Reading Mike Cohn Agile Estimating and Planning provides guidance on iteration planning, including estimating the effort for user stories. David J. Anderson Kanban: Successful Evolutionary Change for Your Technology Business provides the guidance, definitions, and metric calculations necessary to establish an efficient software development flow, including establishing WIP limits.

Software is nothing like construction. Software is innovation. Innovation is difficult—if not impossible—to predict.

And, of course, as savvy readers can appreciate, when software developers say it’s going to take a year to get something done, they really mean two years. It’s not because they’re incompetent, or that they are calendar-challenged, it’s just that estimating the time to do something we’ve never done before is something we suck at. And, by nature, we’re often optimistic animals.

Now I do not think software managers have less inherent courage and firmness than chefs, nor than other engineering managers. But false scheduling to match the patron's desired date is much more common in our discipline than elsewhere in engineering. It is very difficult to make a vigorous, plausible, and job-risking defense of an estimate that is derived by no quantitative method, supported by little data, and certified chiefly by the hunches of the managers. Clearly

In 1982, economists at the Brookings Institute estimated that about 62 per cent of the value of a typical American firm stemmed from its physical assets—everything from tables and chairs to factories and inventories. Everything else consisted of more intangible “knowledge assets.” By 1992, the balance had completely reversed. They calculated that only 38 per cent of the average firm’s value came from its physical assets. With the shift towards more knowledge-intensive production processes, it is natural that firms should start to worry much more about employee loyalty. It is relatively easy to stop employees from making off with company property—just post guards at the gate. But when employees leave, they generally take with them all the knowledge and experience they have acquired, and there is no way to stop them. So the best way for a firm to retain control of its assets is to build a strong organizational culture, one that will inspire loyalty and allegiance from its employees. From this perspective, it is entirely predictable that the firms that depend most heavily on the knowledge of their workers will also be the firms that put the most effort into employee retention. Software companies in particular are famous for their efforts to create a corporate culture that will secure employee allegiance.

Well, yes, software people are terrible at estimating, because humans are terrible at estimating. Let’s not just try harder. Let’s find a better way.

software people are terrible at estimating, because humans are terrible at estimating. Let’s not just try harder. Let’s find a better way.

A good estimate is an estimate that provides a clear enough view of the project reality to allow the project leadership to make good decisions about how to control the project to hit its targets.

Equally, when software developers estimated completion time for 60 tasks over three months, the estimated time varied by an average of 71%.399

staffers in the FAA’s Aircraft Certification Service concluded there might be fifteen more MAX crashes without a software fix, based on the future size of the fleet, the hours of anticipated flight, and a rough estimate that one in every hundred pilots might have trouble handling the rare sensor failure.

underestimations are less likely to be challenged, and cause us more disappointment down the road, than overestimations.

When the automation test pack is being designed, the most important decision is to plan the Test Scheduling of those Automated Test Scripts. The objective of test automation is to reduce the amount of time spent in Regression Testing

In other words, these methodologies don’t change the fact that software engineers are bad at estimating; they just keep the estimates short, so that even a significant slip, in percentage terms, is not that bad

Disinformation Systems consist of elaborate deceptions, constructed by intelligence agencies like the C.I.A., K.G.B. or England's M.I.5, in which a cover story, when created, has within it a second deception, disguised to look like "the hidden truth" to any suspicious rival who successfully digs below the surface. Since Disinformation Systems have multiplied like bacteria in our increasingly clandestine world, any perception psychologist who looks into modern politics will recognize that quantum logic, probability theory and strong doses of zeteticism make the best tools to employ in estimating if the President has just told us another whooping big lie or has just uttered the truth for once.

Making good engineering decisions is all about weighing all of the available inputs and making informed decisions about the trade-offs. Sometimes, those decisions are based on instinct or accepted best practice, but only after we have exhausted approaches that try to measure or estimate the true underlying costs.

As we mentioned in chapter 4, any accounting change that is “material” to the bottom line must be footnoted in this manner. But who decides what is material and what isn’t? You guessed it: the accountants. In fact, it could very well be that recognizing 75 percent up front presents a more accurate picture of the software division’s reality. But was the change in accounting method due to good financial analysis, or did it reflect the need to make the earnings forecast? Could there be a bias lurking in here? Remember, accounting is the art of using limited data to come as close as possible to an accurate description of how well a company is performing. Revenue on the income statement is an estimate, a best guess. This example shows how estimates can introduce bias.

By focusing on key metrics, and worrying less about adherence to estimates, budgets, and plans, tech companies remove a lot of the drama that comes with the traditional project and put all their energy and focus into the work and the customer instead.

the initial design is likely to be the most prosaic and naive implementation. Insight comes from working with the system.

from the data I have I think it will take at least X amount of time, but I can’t give you an absolute certain answer. What I can do is work as diligently as possible, give you frequent updates on my progress and revise the estimations as frequently as you would like.

What can explain this difference? On the surface, much appears to hinge on Richard’s programming feat, his software shim. Otherwise, his effort with Konqueror seems much like my struggles with Mozilla. Perhaps he was just a better programmer than me, and without his coding cleverness, there would be no story. That explanation is too simple. Richard made his shim only after determining he needed one last link in a chain of inspiration, intuition, reasoning, and estimation. His shim was a consequence of his overall plan. To show what I mean, here’s an accounting of what Richard did in his first couple of days at Apple. He began by quizzing us on the browser analysis we had done before his arrival, and after hearing it, he quickly discarded our effort with Mozilla as unlikely to bear fruit. By doing so, he demonstrated the self-confidence to skip any ingratiating display of deference to his new manager, a person who had years of experience in the technical field he was newly entering. Next, Richard resolved to produce a result on the shortest possible schedule. He downloaded an open source project that held genuine promise, the Konqueror code from KDE, a browser that might well serve as the basis for our long-term effort. In getting this code running on a Mac, he decided to make the closest possible approximation of a real browser that was feasible on his short schedule. He identified three features—loading web pages, clicking links, and going back to previous pages. He reasoned these alone would be sufficiently compelling proofs of concept. He then made his shortcuts, and these simplifying choices defined a set of nongoals: Perfect font rendering would be cast aside, as would full integration with the Mac’s native graphics system, same for using only the minimum source code from KDE. He reasoned that these shortcuts, while significant, would not substantially detract from the impact of seeing a browser surf web pages. He resolved to draw together these strands into a single demo that would show the potential of Konqueror. Then, finally, he worked through the technical details, which led him to develop his software shim, since that was the only thing standing between him and the realization of his plan. His thought process amplified his technical acumen. In contrast, Don and I were hoping Mozilla would pan out somehow. I was trying to get the open source behemoth to build on the Mac, with little thought beyond that. I had no comparable plan, goals, nongoals, tight schedule, or technical shortcuts.

Second, when velocity is used as a productivity measure, teams inevitably work to game their velocity. They inflate their estimates and focus on completing as many stories as possible at the expense of collaboration with other teams (which might decrease their velocity and increase the other team’s velocity, making them look bad). Not

The more predictable you are, the better you will be at estimating how long something will take and the more promises you can make about your work.

Very few companies know how to exploit the data already embedded in their core operating systems. THE SOLUTION Evidence-based, data-driven decision making provides the answer, but it requires a big cultural shift and four changes in how operations are managed. Who Benefits from Big Data? 496 words Big data is big business. The IT research firm Gartner estimates that total software, social media, and IT services spending related to big data and analytics topped $28 billion worldwide in 2012. All estimates predict rapid growth. In addition to vendors, at least three types of organizations are harvesting value from big data.