Paul Graham: The Other Side of "Masterpieces on Time"
“Good artists create, great artists steal, and real artists execute an order on time.”
One of the differences between large companies and start-ups is that the former have developed procedures that protect them from errors. Startups are like a one-year-old kid, constantly falling and knocking on corners. Big companies are much more circumspect.
')
The gradual accumulation of checks in an organization can be compared to the learning process of failures that have happened to them or similar companies. For example, after concluding a contract with a supplier that goes bankrupt and does not fulfill its obligations, the company can begin to demand from all suppliers proof of their solvency.
As a company grows, it invariably acquires such checks, either in response to its own failures, or, even more often, by hiring people from larger organizations that bring with them their usual ways of protecting them from new misfortunes.
In the process of learning from mistakes for the company there is nothing unusual. The problem is that people who offer to introduce new checks almost never think about their cost.
Each check has a cost. For example, take the requirement of confirmation by suppliers of their solvency. Sure it's prudent? After all, in fact, it leads to significant costs. Obvious direct loss of time of people from both sides, both from suppliers, and from considering their proofs is obvious. But the real costs are precisely those about which no one will tell you: a company that could become the best supplier did not participate in the tender because of the unwillingness to waste resources, taking into account the possible failure of the inspection. Or that will not pass the threshold value of the solvency test, which, of course, will be greatly overestimated, because the real cost of this test is not so obvious.
Whenever anyone in an organization offers to introduce a new check, let them first explain not only its advantages, but also its cost. And even if its cost cannot be calculated correctly, it will serve as a reminder that it exists and it is advisable to find it out.
If companies do this, they will have surprises. Joel Spolsky recently told Y Combinator about selling software to corporate clients. He said that in most companies the decision to purchase software worth about $ 1,000 is made by managers without any additional approvals. The acquisition of more expensive software usually must be approved by the commissions. And then “hatching out” their buying decisions becomes so expensive for sellers that they have to charge $ 50,000 for their products instead of $ 5,000.
Apparently, the goal of all these commissions is to prevent wasting company money. As a result, the company pays ten times more.
Purchase checks will always be expensive, because the harder you sell, the higher the price. In addition, the companies that specialize in selling such difficult customers, and not professionals of their business, will contact you. And so the unproductiveness is deeper and deeper: market mechanisms work against you - good suppliers are looking for the best customers.
This situation is constantly getting the biggest organizations of all - the government. But the checks used by officials can cause great harm besides overpayment - they can bury the entire economy of the country. Until 1400, China was richer and more technologically advanced than Europe. One of the reasons for the advanced positions of Europe is the ban of the Chinese authorities on the implementation of long trade trips. The Europeans got all the benefits of intelligence and research, they began to dominate the rest of the world, including China.
In less distant times, the Sarbanes-Oxley Act virtually destroyed the initial public offerings (IPO) market in the United States. This is not what the authors of this bill wanted. They just wanted to add some audits to public companies. But they forgot to think about their value. They have forgotten that companies resorting to IPO are for the most part cramped in funds and that the additional burden of checks that General Electric will not notice, for example, may simply deprive them of the opportunity to attract financing in the securities market.
Starting to think about the price of inspections, you might have other interesting questions. Does their cost decrease or increase? Is it higher in some areas than in others? Where is its ceiling? If large organizations start to be interested in these issues, they will find out more than one frightening thing.
I think the cost of inspections may increase. Mainly due to the fact that software plays an ever-increasing role in companies, and checks are especially contraindicated for people who create it.
Programmers are not typical workers - the best of them prefer to work hard. This is not a typical case for other areas of activity. When I worked in the catering, we did not like rush hours. And when I mowed the lawn, I definitely didn’t want the grass to grow faster.
Programmers are more pleasant to write more code. Or, more precisely, to implement. Programmers will understand the difference. At least good ones.
For such, the beauty of work in startups is in the absence of a mass of checks when a product is released. In these startups there are no external checks at all. If in the morning you suddenly had an idea about something new, you can write it and push it to the working servers by lunchtime. And when you have this opportunity, you have more ideas.
In large companies, software products must pass several endorsements before they start. And the cost of this can be enormous, there is no limit. I recently spoke with a group of three programmers whose startup was bought several years ago by a large company. When they were independent, they could make changes instantly. Now, according to them, two weeks is the minimum period for which they can run the new code into operation.
However, this did not make them less productive. They just began to hate employment.
This is what clearly demonstrates the desire of programmers to work hard: these guys would pay for the immediate launch of codes in the work, as they used to do it. When I asked them if they would change 10% of the amount paid on the acquisition, to the possibility of instantly launching new codes into the work, all three immediately agreed. Then I asked, and how much of the maximum percentage of the amount they are willing to pay for it. They said they did not want to even think about it, afraid to go too far, but I feel that they would give up to 50 percent.
Those. they would donate hundreds of thousands of dollars, maybe even millions, simply to supply more software to users. And you know what? This can be trusted with no fear. Even the buyer company would only be better off from this: they would not only not break anything, but, on the contrary, would do much more. It turns out that the buyer actually gets the worst result for a higher price.
As well as the commission for the approval of the acquisition of software.
As a result, if for a company overgrown with barriers to making purchases, the greatest danger lies not in overpayment, but in the reluctance of the best market players to work with this company. There is a similar danger in the abuse of checks when new codes are released - your programmers simply will not want to work with you.
The famous maxim of Steve Jobs “giving a masterpiece on time” (an artists ship) works both ways. Developers can not only finish the job on time. They insist on it. If you prevent them, you will remain without masterpieces at all.
Thank you for the help with the transfer to Sergey Danshin and the Edison company (which did the technical audit of the economic game and the centralized video management system ).
“Just-in-time masterpieces” - translating Steve Jobs's short and deep aphorisms is not an easy task, to say the least. In the original language “masterpieces just in time” fit literally into three words - “Real artists ship”, and the meaning of this phrase varies slightly depending on the context. For example, "Jobs also reminds his employees that real artists ship," and "the fulfillment of plans on time is no less important than innovation and unique design. "). At some point, the brutal “masterpiece, bitch, and so that during,” was taken as the working version of the “Real artists ship” translation, until we came across a sequel to the famous aphorism about artists: “Good artists create, great artists steal, and real artists ship ”-“ good artists create, great artists steal, and real artists perform the order on time ”. As a result, stopped at the "masterpieces just in time."
[ source ]
The Other Half of "Artists Ship"
Paul Graham, November 2008One of the differences between large companies and start-ups is that the former have developed procedures that protect them from errors. Startups are like a one-year-old kid, constantly falling and knocking on corners. Big companies are much more circumspect.
')
The gradual accumulation of checks in an organization can be compared to the learning process of failures that have happened to them or similar companies. For example, after concluding a contract with a supplier that goes bankrupt and does not fulfill its obligations, the company can begin to demand from all suppliers proof of their solvency.
As a company grows, it invariably acquires such checks, either in response to its own failures, or, even more often, by hiring people from larger organizations that bring with them their usual ways of protecting them from new misfortunes.
In the process of learning from mistakes for the company there is nothing unusual. The problem is that people who offer to introduce new checks almost never think about their cost.
Each check has a cost. For example, take the requirement of confirmation by suppliers of their solvency. Sure it's prudent? After all, in fact, it leads to significant costs. Obvious direct loss of time of people from both sides, both from suppliers, and from considering their proofs is obvious. But the real costs are precisely those about which no one will tell you: a company that could become the best supplier did not participate in the tender because of the unwillingness to waste resources, taking into account the possible failure of the inspection. Or that will not pass the threshold value of the solvency test, which, of course, will be greatly overestimated, because the real cost of this test is not so obvious.
Whenever anyone in an organization offers to introduce a new check, let them first explain not only its advantages, but also its cost. And even if its cost cannot be calculated correctly, it will serve as a reminder that it exists and it is advisable to find it out.
If companies do this, they will have surprises. Joel Spolsky recently told Y Combinator about selling software to corporate clients. He said that in most companies the decision to purchase software worth about $ 1,000 is made by managers without any additional approvals. The acquisition of more expensive software usually must be approved by the commissions. And then “hatching out” their buying decisions becomes so expensive for sellers that they have to charge $ 50,000 for their products instead of $ 5,000.
Apparently, the goal of all these commissions is to prevent wasting company money. As a result, the company pays ten times more.
Purchase checks will always be expensive, because the harder you sell, the higher the price. In addition, the companies that specialize in selling such difficult customers, and not professionals of their business, will contact you. And so the unproductiveness is deeper and deeper: market mechanisms work against you - good suppliers are looking for the best customers.
This situation is constantly getting the biggest organizations of all - the government. But the checks used by officials can cause great harm besides overpayment - they can bury the entire economy of the country. Until 1400, China was richer and more technologically advanced than Europe. One of the reasons for the advanced positions of Europe is the ban of the Chinese authorities on the implementation of long trade trips. The Europeans got all the benefits of intelligence and research, they began to dominate the rest of the world, including China.
In less distant times, the Sarbanes-Oxley Act virtually destroyed the initial public offerings (IPO) market in the United States. This is not what the authors of this bill wanted. They just wanted to add some audits to public companies. But they forgot to think about their value. They have forgotten that companies resorting to IPO are for the most part cramped in funds and that the additional burden of checks that General Electric will not notice, for example, may simply deprive them of the opportunity to attract financing in the securities market.
Starting to think about the price of inspections, you might have other interesting questions. Does their cost decrease or increase? Is it higher in some areas than in others? Where is its ceiling? If large organizations start to be interested in these issues, they will find out more than one frightening thing.
I think the cost of inspections may increase. Mainly due to the fact that software plays an ever-increasing role in companies, and checks are especially contraindicated for people who create it.
Programmers are not typical workers - the best of them prefer to work hard. This is not a typical case for other areas of activity. When I worked in the catering, we did not like rush hours. And when I mowed the lawn, I definitely didn’t want the grass to grow faster.
Programmers are more pleasant to write more code. Or, more precisely, to implement. Programmers will understand the difference. At least good ones.
For such, the beauty of work in startups is in the absence of a mass of checks when a product is released. In these startups there are no external checks at all. If in the morning you suddenly had an idea about something new, you can write it and push it to the working servers by lunchtime. And when you have this opportunity, you have more ideas.
In large companies, software products must pass several endorsements before they start. And the cost of this can be enormous, there is no limit. I recently spoke with a group of three programmers whose startup was bought several years ago by a large company. When they were independent, they could make changes instantly. Now, according to them, two weeks is the minimum period for which they can run the new code into operation.
However, this did not make them less productive. They just began to hate employment.
This is what clearly demonstrates the desire of programmers to work hard: these guys would pay for the immediate launch of codes in the work, as they used to do it. When I asked them if they would change 10% of the amount paid on the acquisition, to the possibility of instantly launching new codes into the work, all three immediately agreed. Then I asked, and how much of the maximum percentage of the amount they are willing to pay for it. They said they did not want to even think about it, afraid to go too far, but I feel that they would give up to 50 percent.
Those. they would donate hundreds of thousands of dollars, maybe even millions, simply to supply more software to users. And you know what? This can be trusted with no fear. Even the buyer company would only be better off from this: they would not only not break anything, but, on the contrary, would do much more. It turns out that the buyer actually gets the worst result for a higher price.
As well as the commission for the approval of the acquisition of software.
As a result, if for a company overgrown with barriers to making purchases, the greatest danger lies not in overpayment, but in the reluctance of the best market players to work with this company. There is a similar danger in the abuse of checks when new codes are released - your programmers simply will not want to work with you.
The famous maxim of Steve Jobs “giving a masterpiece on time” (an artists ship) works both ways. Developers can not only finish the job on time. They insist on it. If you prevent them, you will remain without masterpieces at all.
Thank you for the help with the transfer to Sergey Danshin and the Edison company (which did the technical audit of the economic game and the centralized video management system ).
“Just-in-time masterpieces” - translating Steve Jobs's short and deep aphorisms is not an easy task, to say the least. In the original language “masterpieces just in time” fit literally into three words - “Real artists ship”, and the meaning of this phrase varies slightly depending on the context. For example, "Jobs also reminds his employees that real artists ship," and "the fulfillment of plans on time is no less important than innovation and unique design. "). At some point, the brutal “masterpiece, bitch, and so that during,” was taken as the working version of the “Real artists ship” translation, until we came across a sequel to the famous aphorism about artists: “Good artists create, great artists steal, and real artists ship ”-“ good artists create, great artists steal, and real artists perform the order on time ”. As a result, stopped at the "masterpieces just in time."
[ source ]
Source: https://habr.com/ru/post/311476/
All Articles