Lately we've been semi-scrumming in the section. We tried to integrate some of the Scrum practices, while ignoring others (the hard ones, you know, like the ones that require management commitment and stuff like that). This isn't so healthy, I know, since the practices were designed to work together; but we're trying to use a sensible subset, in an
(
Read more... )
חסרונות:
1. לשתף את כל הצוות בתכנון [זה רעיון מצוין מהמון בחינות, אבל] מגדיל הנדסית את
הזמן שהתכנון ייקח, פשוט בגלל הכמות הגדולה של האנשים והדעות.
2. ככל שימי (!) התכנון עוברים, הריכוז והנכונות של הצוות ילכו ויפחתו.
3. יתכן בהחלט שאנחנו עושים משהו שלא לפי הספר, אבל הערכות הזמנים העליזות של סקראם (חמש דקות לדיילי! חצי שעה לפוסט ספרינט ריויו!) פשוט לא עומדות במבחן המציאות.
4. אני עדיין לא רואה שום הגיון בשיטה של סקראם לשתף בקביעת לו"זים גם אנשים שבמובהק לא מבינים במשימה כלשהי ושלא יקבלו את המשימה הנדונה - ואז לבחור את הצעת הימים הבזבזנית ביותר, כלומר דוקא של אלו.
מצד שני - ולא רק כי סרגיי טוען שאני שלילית דיפולטיבית -
5. תכנון כלל-צוותי: כל אחד בצוות יודע בדיוק מה עומד לקרות. גם בתכולות שאין לו נגיעה בהם. ולאחרונה נדמה לי שידע זה שליטה זה אכפתיות זה אנשים מרוצים יותר עם ראש גדול יותר והרבה יותר גאוה במוצר הסופי הטוב יותר.
6. דייליז. אותה נקודה בדיוק.
7. הנוהג לרשום תוצר רצוי מול כל תכולה. עדיין אותה הנקודה.
8. פירוק תכולה חודרני עד לרמה של משימות אטומיות ברורות. (נקודה בסעיף 5)
9. המחשה מטריאלית: לא חשבתי על הנקודה הזו עד שהעלית אותה ונזכרתי שתמיד, אבל תמיד, ניהלתי את הטו-דוז שלי על חתיכת נייר. יש דברים שפשוט לא עוברים מסך.
10. ואוי, כמה אני מקוה שזה ילמד אותי פרק בחישוב לו"זים.
אבל בוא נעבור קודם את מבחן המציאות.
Reply
1. That's a true observation. However, if the way I feel about our new estimates is anywhere close to how they'll turn out, I think the two or three days we've spent on planning will be worth it. Also, I expect the number to drop to one or 1.5 in future sprints, simply because we'll be more familiar with the process.
2. That's true as well, but it sounds like an external problem; meaning, not an inherent flaw in Scrum, but rather a technical difficulty that can be overcome by more waffles. Or something.
3. Dailies should be 15 minutes; that is completely realistic if we stick to what the daily should be like: everyone gets their own 2-3 minutes to talk, and that's it. From what I've heard, the ScrumMaster's standard equipment kit includes a 3-minute hourglass and a water gun. You can guess how they're supposed to be used. Do you think it's impossible to summarize your previous day and your upcoming day in three minutes? Again, this might be a matter of practicing.
4. Agreed. I don't think that Scrum dictates that the longest estimate has to be accepted; everywhere I read it says "everybody flips their card at the same time, and then the team discusses the estimates until a consensus is reached". I might have taken Danko's example a little too literally there. Also, the "everyone must estimate" rule only works for tasks belonging to a common discipline, like development. Design, for example, is shared only by 2.5 persons on the team, and thus there really is no use in forcing the others to estimate design-related tasks.
5. I feel the same way, and I really hope that's the way the rest of the team sees it.
6. Same. We just have to make them a little shorter so no one gets bored.
7. That actually isn't Scrum; that's me [blushes]. I find it's easier for me to estimate and accept tasks when I know what the final product is supposed to be, so I started noting those on the whiteboard to help the others. Glad it works.
8. I just LOVE that. Disadvantage: you can't estimate anything you can't break down. I'm not sure that's a disadvantage, because it only means your estimate would've been off anyway.
9. Agreed. I'm really surprised at myself about this, though, because I used to be a rather purist technologist.
10. As do us all. I hate to break it to you, but the dark art of true estimation is mastered by a select few; us mortals can only try our best - and hope. No doubt breaking tasks down and agreeing on a definite scope helps a lot, though.
Reply
Leave a comment